On 28/02/16 17:43, Chris Young wrote:
> Hi
>
> I've just updated the ppc-amigaos toolchain to use gcc 5.3.
>
> It needs building, if somebody can kick that off for me please. :)
Given that there are multiple issues in 3.4 that render it useless on a
number of platforms, we're intending to spin a 3.5 ASAP. On that basis,
I'd like to hold off on updating the toolchain until after 3.5 is done,
to minimise the opportunity for disappointment.
J.
Monday, 29 February 2016
Sunday, 28 February 2016
ppc-amigaos gcc 5.3
Hi
I've just updated the ppc-amigaos toolchain to use gcc 5.3.
It needs building, if somebody can kick that off for me please. :)
Thanks
Chris
I've just updated the ppc-amigaos toolchain to use gcc 5.3.
It needs building, if somebody can kick that off for me please. :)
Thanks
Chris
Monday, 22 February 2016
Re: We're in a mess
In article Dave Higton <dave@davehigton.me.uk> wrote:
> 1) We're distributing an out of date version of SharedULib. We
> are sending out 1.12; the most recent version is 1.14 (or maybe
> 1.14-1).
While it may not be the latest, it's newer than 1.07 required by the
RMEnsure, so in that sense it's not out of date from NetSurf's point of view.
> 2) The file dates of the resources we ship in !Boot and !System
> bear no relation to reality.
>
> Here's what I propose: [snip]
>
> c) My best suggestion to assist updating of !Boot/!System is
> to provide an app along with the NS archive, which users only
> have to double-click on (where it is, in the archive) to perform
> the updates via the standard Installer tool.
To play devil's advocate: if I'm already Mr.L.Azy user who is ignoring the
!Boot and !System provided in the NetSurf archive, am I just as likely to
ignore some update app too?
In article Rob Kendrick <rjek@netsurf-browser.org> wrote:
> Upgrading !Boot and !System using the OS-provided functionality has been
> the "thing" for installing apps in RISC OS for the best part of 15
> years.
Yes, that seems to be the real cause here. One user has regressed to their
pre-RISC OS 3.80 ways and copied !Boot/!System updates in manually. The
!System one is particularly special since to manually navigate to the
Resources directory requires more mouse clicks than dragging the !System into
SysMerge in !Configure, which would have checked the module versions &
avoided the problem
Perhaps a cheaper solution here is to put the extra bits and bobs in a
directory within the ZIP called "Dependencies" or similar. Then they're not
tempting to fiddle with unnecessarily. Then change/update NetSurf's !Run file
to check for them (it already checks for the modules), so adding an
IfThere !Cache
IfThere !Unicode
or checking for Unicode$Dir and Cache$Dir being set would complete the set,
Sprow.
> 1) We're distributing an out of date version of SharedULib. We
> are sending out 1.12; the most recent version is 1.14 (or maybe
> 1.14-1).
While it may not be the latest, it's newer than 1.07 required by the
RMEnsure, so in that sense it's not out of date from NetSurf's point of view.
> 2) The file dates of the resources we ship in !Boot and !System
> bear no relation to reality.
>
> Here's what I propose: [snip]
>
> c) My best suggestion to assist updating of !Boot/!System is
> to provide an app along with the NS archive, which users only
> have to double-click on (where it is, in the archive) to perform
> the updates via the standard Installer tool.
To play devil's advocate: if I'm already Mr.L.Azy user who is ignoring the
!Boot and !System provided in the NetSurf archive, am I just as likely to
ignore some update app too?
In article Rob Kendrick <rjek@netsurf-browser.org> wrote:
> Upgrading !Boot and !System using the OS-provided functionality has been
> the "thing" for installing apps in RISC OS for the best part of 15
> years.
Yes, that seems to be the real cause here. One user has regressed to their
pre-RISC OS 3.80 ways and copied !Boot/!System updates in manually. The
!System one is particularly special since to manually navigate to the
Resources directory requires more mouse clicks than dragging the !System into
SysMerge in !Configure, which would have checked the module versions &
avoided the problem
Perhaps a cheaper solution here is to put the extra bits and bobs in a
directory within the ZIP called "Dependencies" or similar. Then they're not
tempting to fiddle with unnecessarily. Then change/update NetSurf's !Run file
to check for them (it already checks for the modules), so adding an
IfThere !Cache
IfThere !Unicode
or checking for Unicode$Dir and Cache$Dir being set would complete the set,
Sprow.
Sunday, 21 February 2016
Re: [Rpcemu] Problems with 0.8.14 and RISC OS 3.7 in specific config
In message <1553145895.422034.1456014343807.JavaMail.open-xchange@pati
na.store>
Steffen Huber <steffen@huber-net.de> wrote:
> normally I only use RPCEmu with RO4 and RO5. Today, I tried to use
> it with RO 3.7. A fresh install, putting UniBoot into HostFS. Boot
> fails with a strange
> "Internal Error: undefined instruction at &000A5DBC"
For info, I got the same error when I setup 3.7 on my laptop running
Linux Mint.
IIRC setting it to 0MB VRAM got things going again.
Peter is aware of this as I did it at ROUGOL after his talk last week
while he was standing there!
Bryan.
--
RISC OS User Group Of London - http://www.rougol.jellybaby.net/
RISC OS London Show - http://www.riscoslondonshow.co.uk/
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
na.store>
Steffen Huber <steffen@huber-net.de> wrote:
> normally I only use RPCEmu with RO4 and RO5. Today, I tried to use
> it with RO 3.7. A fresh install, putting UniBoot into HostFS. Boot
> fails with a strange
> "Internal Error: undefined instruction at &000A5DBC"
For info, I got the same error when I setup 3.7 on my laptop running
Linux Mint.
IIRC setting it to 0MB VRAM got things going again.
Peter is aware of this as I did it at ROUGOL after his talk last week
while he was standing there!
Bryan.
--
RISC OS User Group Of London - http://www.rougol.jellybaby.net/
RISC OS London Show - http://www.riscoslondonshow.co.uk/
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: We're in a mess
On Sun, Feb 21, 2016 at 09:13:55PM +0000, Dave Higton wrote:
>
> Comments, please!
Upgrading !Boot and !System using the OS-provided functionality has been
the "thing" for installing apps in RISC OS for the best part of 15
years. If users think they can install stuff by simply dragging
!NetSurf out of a zip they need education. It hasn't been that simple
in *ages*.
If the merge tool in the configure app/!Boot doesn't pay attention to
the version and date information in the file, then the bug is there.
B.
>
> Comments, please!
Upgrading !Boot and !System using the OS-provided functionality has been
the "thing" for installing apps in RISC OS for the best part of 15
years. If users think they can install stuff by simply dragging
!NetSurf out of a zip they need education. It hasn't been that simple
in *ages*.
If the merge tool in the configure app/!Boot doesn't pay attention to
the version and date information in the file, then the bug is there.
B.
Re: Proposed app
In article <2cf0945555.DaveMeUK@my.inbox.com>,
Dave Higton <dave@davehigton.me.uk> wrote:
> It could do with a few more comments. Maybe sprites (anyone think of
> a meaningful picture?)
>
> For review and comments.
In the obey file, RMLoad, not RMRun.
Bonus marks for using OS_GBPB properly!
Sprow.
Dave Higton <dave@davehigton.me.uk> wrote:
> It could do with a few more comments. Maybe sprites (anyone think of
> a meaningful picture?)
>
> For review and comments.
In the obey file, RMLoad, not RMRun.
Bonus marks for using OS_GBPB properly!
Sprow.
Proposed app
It could do with a few more comments. Maybe sprites (anyone think of
a meaningful picture?)
For review and comments.
Dave
____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
Check it out at http://www.inbox.com/earth
a meaningful picture?)
For review and comments.
Dave
____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
Check it out at http://www.inbox.com/earth
We're in a mess
Guys, we're in a mess. You'll be aware if you read RISC OS Open
fora.
I don't think it should take us many hours of work to tidy it up,
though.
1) We're distributing an out of date version of SharedULib. We
are sending out 1.12; the most recent version is 1.14 (or maybe
1.14-1).
2) The file dates of the resources we ship in !Boot and !System
bear no relation to reality.
3) Although we programmers think of updating !Boot and !System
to be easy, ordinary users don't necessarily feel the same way.
Failing to update is widespread; updating by simply copying
new over old is done by some.
The combination of the above, and non-standard (but entirely
understandable) methods of updating !System, have resulted in
at least one user regressing from SharedULib 1.14 to 1.12, which
in turn stopped one of his important apps working.
Here's what I propose:
a) We update SharedULib to the most recent one. I don't know if
there is anything about the most recent one that will stop NS
working correctly - I can do my share of testing, and other RO
users will chip in too. I haven't seen any reports of problems
yet with 1.14, so it looks promising. Of course, if there is
any known problem, we (and I suppose that means me!) have to
work with the SharedULib devs on an agreed solution.
b) We implement the suggestion from IRC the other evening of
touching the file dates to the last git date. The file dates
are a safety feature, and, by changing file dates frequently,
we throw the safety away.
c) My best suggestion to assist updating of !Boot/!System is
to provide an app along with the NS archive, which users only
have to double-click on (where it is, in the archive) to perform
the updates via the standard Installer tool. I've got an app
to offer, in a very nearly finished version, for you to review.
Comments, please!
Dave
____________________________________________________________
Can't remember your password? Do you need a strong and secure password?
Use Password manager! It stores your passwords & protects your account.
Check it out at http://mysecurelogon.com/password-manager
fora.
I don't think it should take us many hours of work to tidy it up,
though.
1) We're distributing an out of date version of SharedULib. We
are sending out 1.12; the most recent version is 1.14 (or maybe
1.14-1).
2) The file dates of the resources we ship in !Boot and !System
bear no relation to reality.
3) Although we programmers think of updating !Boot and !System
to be easy, ordinary users don't necessarily feel the same way.
Failing to update is widespread; updating by simply copying
new over old is done by some.
The combination of the above, and non-standard (but entirely
understandable) methods of updating !System, have resulted in
at least one user regressing from SharedULib 1.14 to 1.12, which
in turn stopped one of his important apps working.
Here's what I propose:
a) We update SharedULib to the most recent one. I don't know if
there is anything about the most recent one that will stop NS
working correctly - I can do my share of testing, and other RO
users will chip in too. I haven't seen any reports of problems
yet with 1.14, so it looks promising. Of course, if there is
any known problem, we (and I suppose that means me!) have to
work with the SharedULib devs on an agreed solution.
b) We implement the suggestion from IRC the other evening of
touching the file dates to the last git date. The file dates
are a safety feature, and, by changing file dates frequently,
we throw the safety away.
c) My best suggestion to assist updating of !Boot/!System is
to provide an app along with the NS archive, which users only
have to double-click on (where it is, in the archive) to perform
the updates via the standard Installer tool. I've got an app
to offer, in a very nearly finished version, for you to review.
Comments, please!
Dave
____________________________________________________________
Can't remember your password? Do you need a strong and secure password?
Use Password manager! It stores your passwords & protects your account.
Check it out at http://mysecurelogon.com/password-manager
Re: Met office missing icons
In message <63441.82.153.33.53.1456075577.squirrel@email.orpheusnet.co.uk>
"Gerald Dodson" <gerald.dodson@argonet.co.uk> wrote:
>> On 21 Feb, "Gerald Dodson" wrote in message
>> <57169.82.153.33.53.1456051649.squirrel@email.orpheusnet.co.uk>:
>>
>>> Recently the mobile forecast has been downloading with the precipitation
>>> icons missing. Everything else is there i.e probability of ?, temp,
>>> feels
>>> like etc. I have looked at the forecast on an android but the site looks
>>> quite different. An earlier NS shows the same lack of icons.
>>>
>>> NS 3421 Iyo 5.18
>>
>> Which weather site is this?
>>
>> --
>> Steve Fryatt - Leeds, England
>
>
>www.metoffice.gov.uk/mobile/forecast/u10qr36jhrgt
Please raise a Mantis case, Gerald.
Dave
____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
Check it out at http://www.inbox.com/earth
"Gerald Dodson" <gerald.dodson@argonet.co.uk> wrote:
>> On 21 Feb, "Gerald Dodson" wrote in message
>> <57169.82.153.33.53.1456051649.squirrel@email.orpheusnet.co.uk>:
>>
>>> Recently the mobile forecast has been downloading with the precipitation
>>> icons missing. Everything else is there i.e probability of ?, temp,
>>> feels
>>> like etc. I have looked at the forecast on an android but the site looks
>>> quite different. An earlier NS shows the same lack of icons.
>>>
>>> NS 3421 Iyo 5.18
>>
>> Which weather site is this?
>>
>> --
>> Steve Fryatt - Leeds, England
>
>
>www.metoffice.gov.uk/mobile/forecast/u10qr36jhrgt
Please raise a Mantis case, Gerald.
Dave
____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
Check it out at http://www.inbox.com/earth
Re: Met office missing icons
> On 21 Feb, "Gerald Dodson" wrote in message
> <57169.82.153.33.53.1456051649.squirrel@email.orpheusnet.co.uk>:
>
>> Recently the mobile forecast has been downloading with the precipitation
>> icons missing. Everything else is there i.e probability of ?, temp,
>> feels
>> like etc. I have looked at the forecast on an android but the site looks
>> quite different. An earlier NS shows the same lack of icons.
>>
>> NS 3421 Iyo 5.18
>
> Which weather site is this?
>
> --
> Steve Fryatt - Leeds, England
www.metoffice.gov.uk/mobile/forecast/u10qr36jhrgt
Gerald
> <57169.82.153.33.53.1456051649.squirrel@email.orpheusnet.co.uk>:
>
>> Recently the mobile forecast has been downloading with the precipitation
>> icons missing. Everything else is there i.e probability of ?, temp,
>> feels
>> like etc. I have looked at the forecast on an android but the site looks
>> quite different. An earlier NS shows the same lack of icons.
>>
>> NS 3421 Iyo 5.18
>
> Which weather site is this?
>
> --
> Steve Fryatt - Leeds, England
www.metoffice.gov.uk/mobile/forecast/u10qr36jhrgt
Gerald
Re: Javascript features list
On Sat, 20 Feb 2016 00:41:08 +0100, Ole Loots wrote:
> Am Freitag, den 19.02.2016, 20:11 +0000 schrieb Peter Slegg:
>
> > I am looking at a survey form and wondered if buttons should work ?
>
>
> When I testet JS builds for Atari three weeks ago, onclick for buttons
> worked well.
It's interesting that it does appear to work on buttons, but not on
<a> links.
Chris
> Am Freitag, den 19.02.2016, 20:11 +0000 schrieb Peter Slegg:
>
> > I am looking at a survey form and wondered if buttons should work ?
>
>
> When I testet JS builds for Atari three weeks ago, onclick for buttons
> worked well.
It's interesting that it does appear to work on buttons, but not on
<a> links.
Chris
Re: Javascript in Atari build
> Date: Sun, 21 Feb 2016 11:57:38 +0000 (GMT)
> From: Peter Slegg <p.slegg@scubadivers.co.uk>
> Subject: Javascript in Atari build
>
> I am using Atari build 3420 and JS seems to be disabled. Paypal
> said it was disabled so I tried setting changing the settings
> because I couldn't remember which is on and off and it can't
> yet be set in the choices dialogue.
>
> enable_javascript:1 { on ? }
> enable_javascript:0 { off ?}
>
>
> http://www.javatester.org/javascript.html
>
> JavaScript IS NOT WORKING in your web browser
>
> Peter
>
JavaScript Âá IS WORKING Âá in your web browser
Ignore me, I've been caught by a quirk I found myself.
If enable_javascript:1 is the last line in Choices then it
has to be followed by a newline.
Peter
> From: Peter Slegg <p.slegg@scubadivers.co.uk>
> Subject: Javascript in Atari build
>
> I am using Atari build 3420 and JS seems to be disabled. Paypal
> said it was disabled so I tried setting changing the settings
> because I couldn't remember which is on and off and it can't
> yet be set in the choices dialogue.
>
> enable_javascript:1 { on ? }
> enable_javascript:0 { off ?}
>
>
> http://www.javatester.org/javascript.html
>
> JavaScript IS NOT WORKING in your web browser
>
> Peter
>
JavaScript Âá IS WORKING Âá in your web browser
Ignore me, I've been caught by a quirk I found myself.
If enable_javascript:1 is the last line in Choices then it
has to be followed by a newline.
Peter
Re: Met office missing icons
On 21 Feb, "Gerald Dodson" wrote in message
<57169.82.153.33.53.1456051649.squirrel@email.orpheusnet.co.uk>:
> Recently the mobile forecast has been downloading with the precipitation
> icons missing. Everything else is there i.e probability of ?, temp, feels
> like etc. I have looked at the forecast on an android but the site looks
> quite different. An earlier NS shows the same lack of icons.
>
> NS 3421 Iyo 5.18
Which weather site is this?
--
Steve Fryatt - Leeds, England
http://www.stevefryatt.org.uk/
<57169.82.153.33.53.1456051649.squirrel@email.orpheusnet.co.uk>:
> Recently the mobile forecast has been downloading with the precipitation
> icons missing. Everything else is there i.e probability of ?, temp, feels
> like etc. I have looked at the forecast on an android but the site looks
> quite different. An earlier NS shows the same lack of icons.
>
> NS 3421 Iyo 5.18
Which weather site is this?
--
Steve Fryatt - Leeds, England
http://www.stevefryatt.org.uk/
Javascript in Atari build
> Date: Fri, 19 Feb 2016 21:29:45 +0000
> From: Michael Drake <tlsa@netsurf-browser.org>
> Subject: Re: Javascript features list
> To: netsurf-users@netsurf-browser.org
>
> On 19/02/16 20:11, Peter Slegg wrote:
> >
> > Is there a maintained list of available JS features ?
>
> The closest is this:
>
> http://git.netsurf-browser.org/netsurf.git/tree/Docs/UnimplementedJavascript.txt
>
I am using Atari build 3420 and JS seems to be disabled. Paypal
said it was disabled so I tried setting changing the settings
because I couldn't remember which is on and off and it can't
yet be set in the choices dialogue.
enable_javascript:1 { on ? }
enable_javascript:0 { off ?}
http://www.javatester.org/javascript.html
JavaScript IS NOT WORKING in your web browser
Peter
> From: Michael Drake <tlsa@netsurf-browser.org>
> Subject: Re: Javascript features list
> To: netsurf-users@netsurf-browser.org
>
> On 19/02/16 20:11, Peter Slegg wrote:
> >
> > Is there a maintained list of available JS features ?
>
> The closest is this:
>
> http://git.netsurf-browser.org/netsurf.git/tree/Docs/UnimplementedJavascript.txt
>
I am using Atari build 3420 and JS seems to be disabled. Paypal
said it was disabled so I tried setting changing the settings
because I couldn't remember which is on and off and it can't
yet be set in the choices dialogue.
enable_javascript:1 { on ? }
enable_javascript:0 { off ?}
http://www.javatester.org/javascript.html
JavaScript IS NOT WORKING in your web browser
Peter
Met office missing icons
Recently the mobile forecast has been downloading with the precipitation
icons missing. Everything else is there i.e probability of ?, temp, feels
like etc. I have looked at the forecast on an android but the site looks
quite different. An earlier NS shows the same lack of icons.
NS 3421 Iyo 5.18
icons missing. Everything else is there i.e probability of ?, temp, feels
like etc. I have looked at the forecast on an android but the site looks
quite different. An earlier NS shows the same lack of icons.
NS 3421 Iyo 5.18
Saturday, 20 February 2016
[Rpcemu] Problems with 0.8.14 and RISC OS 3.7 in specific config
RPCEmu 0.8.14 [DYNAREC NO_DEBUG]
Build: 32-bit binary
Compiler: GCC version 4.6.2
OS: Microsoft Windows
OS: PlatformId = 2
OS: MajorVersion = 6
OS: MinorVersion = 1
OS: ProductType = 1
OS: SuiteMask = 0x100
OS: ServicePackMajor = 1
OS: ServicePackMinor = 0
OS: ProcessorArchitecture = 9
OS: SystemMetricsServerR2 = 0
OS: ProductInfoType = 1
Allegro version ID: Allegro 4.4.2, MinGW32
Desktop Resolution: 1920 x 1080
Host Colour Depth: 32
Working Directory: C:\RISCOS\RPCEmu\V0.8.14-spoon\RPCEmuRO3_7
loadconfig: model = "RPCSA"
loadconfig: mouse_twobutton = "0"
loadconfig: network_type = "off"
loadconfig: mouse_following = "1"
loadconfig: cdrom_type = "0"
loadconfig: cdrom_enabled = "1"
loadconfig: refresh_rate = "60"
loadconfig: stretch_mode = "1"
loadconfig: sound_enabled = "0"
loadconfig: vram_size = "2"
loadconfig: mem_size = "128"
loadconfig: ipaddress = "172.31.0.1"
loadconfig: cpu_idle = "0"
romload: Loaded 'ROM370' 4194304 bytes
romload: Total ROM size 4 MB
romload: ROM patch applied: 8MB VRAM RISC OS 3.70
initpodulerom: Successfully loaded 'hostfs,ffa' into podulerom
initpodulerom: Successfully loaded 'hostfsfiler,ffa' into podulerom
initpodulerom: Successfully loaded 'SyncClock,ffa' into podulerom
RPCEmu: Machine reset
RPCEmu: Machine reset complete
HostFS: Registration request version 3 accepted
Hi,
normally I only use RPCEmu with RO4 and RO5. Today, I tried to use
it with RO 3.7. A fresh install, putting UniBoot into HostFS. Boot
fails with a strange
"Internal Error: undefined instruction at &000A5DBC"
After that, not even !Edit runs (abort on data transfer).
Config that fails: StrongARM, 2 MB VRAM, 128 MB RAM.
0.8.13 works fine with the same config.
0.8.14 works fine with RO4 and RO5 with the same config.
Changing the config to e.g. 64 MB RAM or ARM610, 0.8.14 starts working.
I have not tried all permutations :-)
rpclog attached.
Steffen
--
Steffen Huber LambdaComm System - Welcome to Trollinger Country
steffen@huber-net.de
Private homepage http://www.huber-net.de/
RISC OS Blog http://riscosblog.huber-net.de/
Build: 32-bit binary
Compiler: GCC version 4.6.2
OS: Microsoft Windows
OS: PlatformId = 2
OS: MajorVersion = 6
OS: MinorVersion = 1
OS: ProductType = 1
OS: SuiteMask = 0x100
OS: ServicePackMajor = 1
OS: ServicePackMinor = 0
OS: ProcessorArchitecture = 9
OS: SystemMetricsServerR2 = 0
OS: ProductInfoType = 1
Allegro version ID: Allegro 4.4.2, MinGW32
Desktop Resolution: 1920 x 1080
Host Colour Depth: 32
Working Directory: C:\RISCOS\RPCEmu\V0.8.14-spoon\RPCEmuRO3_7
loadconfig: model = "RPCSA"
loadconfig: mouse_twobutton = "0"
loadconfig: network_type = "off"
loadconfig: mouse_following = "1"
loadconfig: cdrom_type = "0"
loadconfig: cdrom_enabled = "1"
loadconfig: refresh_rate = "60"
loadconfig: stretch_mode = "1"
loadconfig: sound_enabled = "0"
loadconfig: vram_size = "2"
loadconfig: mem_size = "128"
loadconfig: ipaddress = "172.31.0.1"
loadconfig: cpu_idle = "0"
romload: Loaded 'ROM370' 4194304 bytes
romload: Total ROM size 4 MB
romload: ROM patch applied: 8MB VRAM RISC OS 3.70
initpodulerom: Successfully loaded 'hostfs,ffa' into podulerom
initpodulerom: Successfully loaded 'hostfsfiler,ffa' into podulerom
initpodulerom: Successfully loaded 'SyncClock,ffa' into podulerom
RPCEmu: Machine reset
RPCEmu: Machine reset complete
HostFS: Registration request version 3 accepted
Hi,
normally I only use RPCEmu with RO4 and RO5. Today, I tried to use
it with RO 3.7. A fresh install, putting UniBoot into HostFS. Boot
fails with a strange
"Internal Error: undefined instruction at &000A5DBC"
After that, not even !Edit runs (abort on data transfer).
Config that fails: StrongARM, 2 MB VRAM, 128 MB RAM.
0.8.13 works fine with the same config.
0.8.14 works fine with RO4 and RO5 with the same config.
Changing the config to e.g. 64 MB RAM or ARM610, 0.8.14 starts working.
I have not tried all permutations :-)
rpclog attached.
Steffen
--
Steffen Huber LambdaComm System - Welcome to Trollinger Country
steffen@huber-net.de
Private homepage http://www.huber-net.de/
RISC OS Blog http://riscosblog.huber-net.de/
Refresh button (Atari)
Atari build 3420
I just had a page timeout (bing) and I noticed that after the
timeout dialogue has been dismissed it isn't possible to click
the refresh button, it is stil greyed out.
Peter
I just had a page timeout (bing) and I noticed that after the
timeout dialogue has been dismissed it isn't possible to click
the refresh button, it is stil greyed out.
Peter
Re: Javascript features list
>Date: Sat, 20 Feb 2016 00:41:08 +0100
>From: Ole Loots <ole@monochrom.net>
>Subject: Re: Javascript features list
>Am Freitag, den 19.02.2016, 20:11 +0000 schrieb Peter Slegg:
>
>> I am looking at a survey form and wondered if buttons should work ?
>
>
>When I tested JS builds for Atari three weeks ago, onclick for buttons
>worked well. What does the verbose log say? And what do you expect?
>
>The code inside the onclick attribute seems to disable the button when
>validation fails. But I guess the rendering engine does not deal with
>such dynamic changes right now. Maybe there is also an exception... the
>log will tell you more.
>
>Greets,
>Ole
Hi Ole,
When I click on the button:
(2332.005000) atari/rootwin.c:176 handle_event: rootwin MU_BUTTON input, x: 674, y: 674
(2332.010003) render/html.c:89 fire_dom_event: Dispatching 'click' against 0x1d21f0c
(2336.240000) atari/gui.c:153 atari_poll: WM: 30
I am expecting NS to go to the next page to begin the survey.
Peter
>From: Ole Loots <ole@monochrom.net>
>Subject: Re: Javascript features list
>Am Freitag, den 19.02.2016, 20:11 +0000 schrieb Peter Slegg:
>
>> I am looking at a survey form and wondered if buttons should work ?
>
>
>When I tested JS builds for Atari three weeks ago, onclick for buttons
>worked well. What does the verbose log say? And what do you expect?
>
>The code inside the onclick attribute seems to disable the button when
>validation fails. But I guess the rendering engine does not deal with
>such dynamic changes right now. Maybe there is also an exception... the
>log will tell you more.
>
>Greets,
>Ole
Hi Ole,
When I click on the button:
(2332.005000) atari/rootwin.c:176 handle_event: rootwin MU_BUTTON input, x: 674, y: 674
(2332.010003) render/html.c:89 fire_dom_event: Dispatching 'click' against 0x1d21f0c
(2336.240000) atari/gui.c:153 atari_poll: WM: 30
I am expecting NS to go to the next page to begin the survey.
Peter
Re: [Rpcemu] RPCEmu Mac OS X test build of 0.8.14
David Pitt, on 19 Feb, wrote:
> Tim Powys-Lybbe, on 19 Feb, wrote:
[snip]
> > Let's see how the odd application works, including ArtWorks.
>
> My initial impressions are good, the real test is to build a ROM, this has
> never completed here on previous versions.
The unpacking did not complete on RPCEmu, not sure why ATM, possibly the
RPCEmu HostFS is gibbing at a filename somewhere. The unpacking was done on
VRPC.
Good news is that a ROM build with !Builder did complete.
> > Then a bit of networking, which I have had working before on a MAC but
> > not for a few years.
>
> Previously networking has been a real stiffer and so abandoned, but I see
> a later TunTap is now available so that will be worth a look at.
>
> http://tuntaposx.sourceforge.net/download.xhtml
This is not good news!
iMac13:~ imac13$ sudo
/Users/imac13/RPCEmu/0.8.14-caliston-dev1/scripts/macosx/startrpcemunet en1
net.inet.ip.forwarding: 1 -> 1
/Users/imac13/RPCEmu/0.8.14-caliston-dev1/scripts/macosx/startnat: line 16:
natd: command not found
/Users/imac13/RPCEmu/0.8.14-caliston-dev1/scripts/macosx/startnat: line 17:
ipfw: command not found
/Users/imac13/RPCEmu/0.8.14-caliston-dev1/scripts/macosx/startnat: line 18:
ipfw: command not found
iMac13:~ imac13$ natd
-bash: natd: command not found
iMac13:~ imac13$ ipfw
-bash: ipfw: command not found
iMac13:~ imac13$
https://developer.apple.com/library/mac/documentation/Darwin/Reference/ManPages/man8/ipfw.8.html
https://developer.apple.com/library/mac/documentation/Darwin/Reference/ManPages/man8/natd.8.html
DEPRECATED. Please use pfctl(8) instead.
DEPRECATED and GONE!!
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
> Tim Powys-Lybbe, on 19 Feb, wrote:
[snip]
> > Let's see how the odd application works, including ArtWorks.
>
> My initial impressions are good, the real test is to build a ROM, this has
> never completed here on previous versions.
The unpacking did not complete on RPCEmu, not sure why ATM, possibly the
RPCEmu HostFS is gibbing at a filename somewhere. The unpacking was done on
VRPC.
Good news is that a ROM build with !Builder did complete.
> > Then a bit of networking, which I have had working before on a MAC but
> > not for a few years.
>
> Previously networking has been a real stiffer and so abandoned, but I see
> a later TunTap is now available so that will be worth a look at.
>
> http://tuntaposx.sourceforge.net/download.xhtml
This is not good news!
iMac13:~ imac13$ sudo
/Users/imac13/RPCEmu/0.8.14-caliston-dev1/scripts/macosx/startrpcemunet en1
net.inet.ip.forwarding: 1 -> 1
/Users/imac13/RPCEmu/0.8.14-caliston-dev1/scripts/macosx/startnat: line 16:
natd: command not found
/Users/imac13/RPCEmu/0.8.14-caliston-dev1/scripts/macosx/startnat: line 17:
ipfw: command not found
/Users/imac13/RPCEmu/0.8.14-caliston-dev1/scripts/macosx/startnat: line 18:
ipfw: command not found
iMac13:~ imac13$ natd
-bash: natd: command not found
iMac13:~ imac13$ ipfw
-bash: ipfw: command not found
iMac13:~ imac13$
https://developer.apple.com/library/mac/documentation/Darwin/Reference/ManPages/man8/ipfw.8.html
https://developer.apple.com/library/mac/documentation/Darwin/Reference/ManPages/man8/natd.8.html
DEPRECATED. Please use pfctl(8) instead.
DEPRECATED and GONE!!
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Friday, 19 February 2016
Re: Javascript features list
Am Freitag, den 19.02.2016, 20:11 +0000 schrieb Peter Slegg:
> I am looking at a survey form and wondered if buttons should work ?
When I testet JS builds for Atari three weeks ago, onclick for buttons
worked well. What does the verbose log say? And what do you expect?
The code inside the onclick attribute seems to disable the button when
validation fails. But I guess the rendering engine does not deal with
such dynamic changes right now. Maybe there is also an exception... the
log will tell you more.
Greets,
Ole
> I am looking at a survey form and wondered if buttons should work ?
When I testet JS builds for Atari three weeks ago, onclick for buttons
worked well. What does the verbose log say? And what do you expect?
The code inside the onclick attribute seems to disable the button when
validation fails. But I guess the rendering engine does not deal with
such dynamic changes right now. Maybe there is also an exception... the
log will tell you more.
Greets,
Ole
Re: Updating !Boot and !System
In article <80b18a5455.DaveMeUK@my.inbox.com>,
Dave Higton <dave@davehigton.me.uk> wrote:
> In message <55547cd825webpages@sprow.co.uk>
> Sprow <webpages@sprow.co.uk> wrote:
> >How many files are present in the !Boot/!System updates from NetSurf?
> >CI 3419 has
> > 13 in !Cache
> > 5 in !System
> > 78 in !Unicode
> >so it's not beyond the wit of man to have 91 lines in an obey file,
> >assuming you use some tool to get their leaf names (or, I suppose, a bit
> >of perl or whatever to generate the script by the build system).
>
> Today I've been writing a small BASIC prog to recurse through all
> the files in a folder and use Install_Update to update the copy
> in a similar location within Boot$Dir or System$Dir as appropriate.
> It's not crunched and is only about 60 lines. It does rely on the
> relative path in the real Boot$Dir/System$Dir being identical to
> that in the !Boot/!System that we provide - that must be so,
> mustn't it?
Yes, the layout of a merge-able !Boot or !System only works if it's the same
structure as the target. In the 2 examples I pointed at it's of the form
Install_Update <Obey$Dir>.Thing <System$Dir>.Thing
> It still doesn't compare the file contents, so, just like the
> pure Copy process, it will typically update more often than needed.
For speed you could just operate a heirarchy
=> Are the date stamps the same?
==> Are the sizes the same?
===> Are the contents the same?
The last one will be slowest, but 1 & 2 will probably catch most files which
are likely to be identical.
> I've written some multi-tasking comparison code elsewhere, and
> I'll see if I can build it into the above tool.
I'm not sure where why "multi-tasking" is specifically mentioned here. It's
only 91 files, if you want it to multi-task make it a TaskObey file.
> >> And it still copies based on the file date (which we know we
> >> can't trust) and not the file content.
> >
> >Not so. See section 5.2 of the spec [...]
>
> Thank you for that... more info that I didn't even know existed!
You might even spot that the other bits of the spec (for example relating to
phrasing of help) got adopted into the Style Guide - the information's out
there, just moving at treacle speed,
Sprow.
Dave Higton <dave@davehigton.me.uk> wrote:
> In message <55547cd825webpages@sprow.co.uk>
> Sprow <webpages@sprow.co.uk> wrote:
> >How many files are present in the !Boot/!System updates from NetSurf?
> >CI 3419 has
> > 13 in !Cache
> > 5 in !System
> > 78 in !Unicode
> >so it's not beyond the wit of man to have 91 lines in an obey file,
> >assuming you use some tool to get their leaf names (or, I suppose, a bit
> >of perl or whatever to generate the script by the build system).
>
> Today I've been writing a small BASIC prog to recurse through all
> the files in a folder and use Install_Update to update the copy
> in a similar location within Boot$Dir or System$Dir as appropriate.
> It's not crunched and is only about 60 lines. It does rely on the
> relative path in the real Boot$Dir/System$Dir being identical to
> that in the !Boot/!System that we provide - that must be so,
> mustn't it?
Yes, the layout of a merge-able !Boot or !System only works if it's the same
structure as the target. In the 2 examples I pointed at it's of the form
Install_Update <Obey$Dir>.Thing <System$Dir>.Thing
> It still doesn't compare the file contents, so, just like the
> pure Copy process, it will typically update more often than needed.
For speed you could just operate a heirarchy
=> Are the date stamps the same?
==> Are the sizes the same?
===> Are the contents the same?
The last one will be slowest, but 1 & 2 will probably catch most files which
are likely to be identical.
> I've written some multi-tasking comparison code elsewhere, and
> I'll see if I can build it into the above tool.
I'm not sure where why "multi-tasking" is specifically mentioned here. It's
only 91 files, if you want it to multi-task make it a TaskObey file.
> >> And it still copies based on the file date (which we know we
> >> can't trust) and not the file content.
> >
> >Not so. See section 5.2 of the spec [...]
>
> Thank you for that... more info that I didn't even know existed!
You might even spot that the other bits of the spec (for example relating to
phrasing of help) got adopted into the Style Guide - the information's out
there, just moving at treacle speed,
Sprow.
Re: [Rpcemu] RPCEmu Mac OS X test build of 0.8.14
Tim Powys-Lybbe, on 19 Feb, wrote:
[snip]
> The next thing will be to find out how to get the Menu key working on this
> trackerball. Currently I can achieve the Menus by doint Alt-Select.
An answer to that would be good! Just to be greedy an adjust click would be
nice.
> Let's see how the odd application works, including ArtWorks.
My initial impressions are good, the real test is to build a ROM, this has
never completed here on previous versions.
> Then a bit of networking, which I have had working before on a MAC but not
> for a few years.
Previously networking has been a real stiffer and so abandoned, but I see a
later TunTap is now available so that will be worth a look at.
http://tuntaposx.sourceforge.net/download.xhtml
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
[snip]
> The next thing will be to find out how to get the Menu key working on this
> trackerball. Currently I can achieve the Menus by doint Alt-Select.
An answer to that would be good! Just to be greedy an adjust click would be
nice.
> Let's see how the odd application works, including ArtWorks.
My initial impressions are good, the real test is to build a ROM, this has
never completed here on previous versions.
> Then a bit of networking, which I have had working before on a MAC but not
> for a few years.
Previously networking has been a real stiffer and so abandoned, but I see a
later TunTap is now available so that will be worth a look at.
http://tuntaposx.sourceforge.net/download.xhtml
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: [Rpcemu] RPCEmu Mac OS X test build of 0.8.14
> On 19 Feb 2016, at 4:50 pm, David Pitt <pittdj@pittdj.co.uk> wrote:
>
> Tim Powys-Lybbe, on 19 Feb, wrote:
>
>>
>>> On 15 Feb 2016, at 11:55 pm, Theo Markettos <theo@markettos.org.uk>
>>> wrote:
>>>
>>> I've made a test build of 0.8.14 for Mac, which you can find at:
>>> https://bitbucket.org/caliston/rpcemu-spoon-caliston/downloads
>
> [snip]
>
>> This time (I failed with 0.8.13) I read the small print very closely and
>> have managed to get RISC running in a microscopic window of about 2 by 1.5
>> inches.
>
> Retina??
No but it is a 4K TV
>
> If so then ‘Get Info’ on RPCEmu.app and tick “Open in Low Resolutionâ€.
Did that and immediately beneficial; I can now read the Emulator easily and without a magnifying glass.
>
>> I had to delete the RPCEmu.plist file and then reboot to get this far.
>>
>> However this micro-window read as:
>>
>> ARM 610 Processor
>>
>> Acorn ADFS
>>
>> Error: Use *Desktop to start TaskManager (Error number &81F403)
>>
>> So I typed in <* Desktop> which produced a micro desktop with this
>> message:
>>
>> Machine startup not completed successfully: 'Disc drive not known'
>
> *co. FileSystem HostFS
That definitely did something for the better but I also wondered why I did not have it all OK from the start. I went to the ROOL site and found Theo's new instructions on getting RPCEmu going on a Mac at <https://www.riscosopen.org/wiki/documentation/show/RPCEmu%20and%20RISC%20OS%205%20on%20Mac%20OS%20X>. So I deleted the Emu data area and plist again and started afresh and it all now works. It still runs at close to 1000 MIPS and on the Mac Activity Monitor the RPCEmu is taking 101% of the MAC's CPU! So I've set the flag on RPCEmu's Preferences to Reduce CPU usage when idle. The Mac's CPU usage is now down to 8% when idling though still top of the list; MIPS are now around 100 and I'll be interested to see what happens when I get the RPCEmu to do some work.
The next thing will be to find out how to get the Menu key working on this trackerball. Currently I can achieve the Menus by doint Alt-Select.
Let's see how the odd application works, including ArtWorks.
Then a bit of networking, which I have had working before on a MAC but not for a few years.
Many thanks for the help.
Tim
--
Tim Powys-Lybbe tim@powys.org
for a miscellany of bygones: http://powys.org/
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
>
> Tim Powys-Lybbe, on 19 Feb, wrote:
>
>>
>>> On 15 Feb 2016, at 11:55 pm, Theo Markettos <theo@markettos.org.uk>
>>> wrote:
>>>
>>> I've made a test build of 0.8.14 for Mac, which you can find at:
>>> https://bitbucket.org/caliston/rpcemu-spoon-caliston/downloads
>
> [snip]
>
>> This time (I failed with 0.8.13) I read the small print very closely and
>> have managed to get RISC running in a microscopic window of about 2 by 1.5
>> inches.
>
> Retina??
No but it is a 4K TV
>
> If so then ‘Get Info’ on RPCEmu.app and tick “Open in Low Resolutionâ€.
Did that and immediately beneficial; I can now read the Emulator easily and without a magnifying glass.
>
>> I had to delete the RPCEmu.plist file and then reboot to get this far.
>>
>> However this micro-window read as:
>>
>> ARM 610 Processor
>>
>> Acorn ADFS
>>
>> Error: Use *Desktop to start TaskManager (Error number &81F403)
>>
>> So I typed in <* Desktop> which produced a micro desktop with this
>> message:
>>
>> Machine startup not completed successfully: 'Disc drive not known'
>
> *co. FileSystem HostFS
That definitely did something for the better but I also wondered why I did not have it all OK from the start. I went to the ROOL site and found Theo's new instructions on getting RPCEmu going on a Mac at <https://www.riscosopen.org/wiki/documentation/show/RPCEmu%20and%20RISC%20OS%205%20on%20Mac%20OS%20X>. So I deleted the Emu data area and plist again and started afresh and it all now works. It still runs at close to 1000 MIPS and on the Mac Activity Monitor the RPCEmu is taking 101% of the MAC's CPU! So I've set the flag on RPCEmu's Preferences to Reduce CPU usage when idle. The Mac's CPU usage is now down to 8% when idling though still top of the list; MIPS are now around 100 and I'll be interested to see what happens when I get the RPCEmu to do some work.
The next thing will be to find out how to get the Menu key working on this trackerball. Currently I can achieve the Menus by doint Alt-Select.
Let's see how the odd application works, including ArtWorks.
Then a bit of networking, which I have had working before on a MAC but not for a few years.
Many thanks for the help.
Tim
--
Tim Powys-Lybbe tim@powys.org
for a miscellany of bygones: http://powys.org/
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: Javascript features list
On 19/02/16 20:11, Peter Slegg wrote:
>
> Is there a maintained list of available JS features ?
The closest is this:
http://git.netsurf-browser.org/netsurf.git/tree/Docs/UnimplementedJavascript.txt
--
Michael Drake http://www.netsurf-browser.org/
>
> Is there a maintained list of available JS features ?
The closest is this:
http://git.netsurf-browser.org/netsurf.git/tree/Docs/UnimplementedJavascript.txt
--
Michael Drake http://www.netsurf-browser.org/
Re: Updating !Boot and !System
In message <55547cd825webpages@sprow.co.uk>
Sprow <webpages@sprow.co.uk> wrote:
>In article <1016645455.DaveMeUK@my.inbox.com>,
> Dave Higton <dave@davehigton.me.uk> wrote:
>> In message <55545d0a67chris@chris-johnson.org.uk>
>> cj <chris@chris-johnson.org.uk> wrote:
>> >In article <8206585455.DaveMeUK@my.inbox.com>,
>> > Dave Higton <dave@davehigton.me.uk> wrote:
>> >> What is this "installer module"?
>> >
>> >*h. installer
>> >==> Help on keyword Installer
>> >Module is: Installer 0.13 (19 Jul 2012)
>>
>> It still looks like a struggle to use it, though. According to
>> the help, it can only cope with a single file per command, and
>> it has to be given both the source path (which we know) and the
>> destination path.
>
>How many files are present in the !Boot/!System updates from NetSurf?
>CI 3419 has
> 13 in !Cache
> 5 in !System
> 78 in !Unicode
>so it's not beyond the wit of man to have 91 lines in an obey file, assuming
>you use some tool to get their leaf names (or, I suppose, a bit of perl or
>whatever to generate the script by the build system).
Today I've been writing a small BASIC prog to recurse through all
the files in a folder and use Install_Update to update the copy
in a similar location within Boot$Dir or System$Dir as appropriate.
It's not crunched and is only about 60 lines. It does rely on the
relative path in the real Boot$Dir/System$Dir being identical to
that in the !Boot/!System that we provide - that must be so,
mustn't it? Currently it doesn't attempt to add folders that don't
already exist in the destination, so it would fail if we added a
new folder. I think this must be a very rare event. It wouldn't
be difficult to support it if required.
It still doesn't compare the file contents, so, just like the
pure Copy process, it will typically update more often than needed.
I've written some multi-tasking comparison code elsewhere, and
I'll see if I can build it into the above tool.
>> And it still copies based on the file date (which we know we
>> can't trust) and not the file content.
>
>Not so. See section 5.2 of the spec
> http://www.marutan.net/wikiref/Acorn%20Registered%20Developer%20REFERNC/RO4/API/HTML/CONFIGUR.HTM
Thank you for that... more info that I didn't even know existed!
However... most of our files are not modules, and not exceptions
as described in the doc, so they're updated on the basis of file
date. The modules are based on version number, and when we put
out a new version, they will have a recent date... so it amounts
to the same thing. When we just put out the same version number
but a later file date, which is what happens rather a lot, they
/will/ be updated on the basis of file date. At least we won't
be bothered by the clashes that exist between the ROL and ROOL
modules...
Dave
____________________________________________________________
Can't remember your password? Do you need a strong and secure password?
Use Password manager! It stores your passwords & protects your account.
Check it out at http://mysecurelogon.com/password-manager
Sprow <webpages@sprow.co.uk> wrote:
>In article <1016645455.DaveMeUK@my.inbox.com>,
> Dave Higton <dave@davehigton.me.uk> wrote:
>> In message <55545d0a67chris@chris-johnson.org.uk>
>> cj <chris@chris-johnson.org.uk> wrote:
>> >In article <8206585455.DaveMeUK@my.inbox.com>,
>> > Dave Higton <dave@davehigton.me.uk> wrote:
>> >> What is this "installer module"?
>> >
>> >*h. installer
>> >==> Help on keyword Installer
>> >Module is: Installer 0.13 (19 Jul 2012)
>>
>> It still looks like a struggle to use it, though. According to
>> the help, it can only cope with a single file per command, and
>> it has to be given both the source path (which we know) and the
>> destination path.
>
>How many files are present in the !Boot/!System updates from NetSurf?
>CI 3419 has
> 13 in !Cache
> 5 in !System
> 78 in !Unicode
>so it's not beyond the wit of man to have 91 lines in an obey file, assuming
>you use some tool to get their leaf names (or, I suppose, a bit of perl or
>whatever to generate the script by the build system).
Today I've been writing a small BASIC prog to recurse through all
the files in a folder and use Install_Update to update the copy
in a similar location within Boot$Dir or System$Dir as appropriate.
It's not crunched and is only about 60 lines. It does rely on the
relative path in the real Boot$Dir/System$Dir being identical to
that in the !Boot/!System that we provide - that must be so,
mustn't it? Currently it doesn't attempt to add folders that don't
already exist in the destination, so it would fail if we added a
new folder. I think this must be a very rare event. It wouldn't
be difficult to support it if required.
It still doesn't compare the file contents, so, just like the
pure Copy process, it will typically update more often than needed.
I've written some multi-tasking comparison code elsewhere, and
I'll see if I can build it into the above tool.
>> And it still copies based on the file date (which we know we
>> can't trust) and not the file content.
>
>Not so. See section 5.2 of the spec
> http://www.marutan.net/wikiref/Acorn%20Registered%20Developer%20REFERNC/RO4/API/HTML/CONFIGUR.HTM
Thank you for that... more info that I didn't even know existed!
However... most of our files are not modules, and not exceptions
as described in the doc, so they're updated on the basis of file
date. The modules are based on version number, and when we put
out a new version, they will have a recent date... so it amounts
to the same thing. When we just put out the same version number
but a later file date, which is what happens rather a lot, they
/will/ be updated on the basis of file date. At least we won't
be bothered by the clashes that exist between the ROL and ROOL
modules...
Dave
____________________________________________________________
Can't remember your password? Do you need a strong and secure password?
Use Password manager! It stores your passwords & protects your account.
Check it out at http://mysecurelogon.com/password-manager
[Rpcemu] Large files in HostFS was Re. RPCEmu Mac OS X test build of 0.8.14
On Tue, Feb 16, 2016 at 05:52:36PM +0000, Sprow wrote:
> In article <20160216173419.GA20045@spod.org>,
> > For the long explanation of why this HostFS patch is not included in
> > RPCEmu, please see this post from 2011.
> >
> > http://www.riscos.info/pipermail/rpcemu/2011-October/001383.html
>
> I recall that discussion at the time, and the thread continued
> http://www.riscos.info/pipermail/rpcemu/2011-October/001384.html
> then fell silent.
>
> The tests performed in the post you've highlighted didn't really elicit any
> new information:
> Old FileSwitch + 4GB files + vanilla HostFS => doesn't work
> Old FileSwitch + 4GB files + 4GB HostFS => doesn't work
> New FileSwitch + 4GB files + vanilla HostFS => doesn't work
> New FileSwitch + 4GB files + 4GB HostFS => does work
>
> the point being that having the underlying HostFS being 4GB capable is
> harmless for OS versions that it already doesn't work on, it's only when
> coupled with RISC OS 5.20 and (chronologically) later that you get benefit.
Unfortuanately it is not harmless on OSes < 5.20 nor harmless on >= 5.20.
At this time it seems to be important to state the requirements that
HostFS should have. These may only have been implicit before.
1) HostFS needs to work on all versions of RISC OS that RPCEmu can run (at
least 3.5, possibly even 3.1 for arcem support).
2) RPCEmu users can place any size file in their HostFS directory on the
Host Side, and HostFS must handle this gracefully.
3) HostFS should protect RPCEmu user's data from loss or corruption.
One specific case discussed here of 3) is that allowing files to be opened
that are larger than the maximum size that RO supports risks data loss, as
a program can only work on only some of the data in a file, which can cause
corruption.
Here follows a table describing the current situation in RPCEmu.
I have used the phrase 'Data Safe' to represent that they don't allow
files to be opened that are larger than OS supports.
32bit builds 64bit builds
RO 3,4,6 Data Safe (2GB filesize limit) Data Unsafe (no filesize limit)
RO 5 Data Safe (2GB filesize limit) Data Unsafe (no filesize limit)
The problem of RPCEmu as it stands at the moment is that it does not meet
our requirements of protecting users data on 64bit builds. Because HostFS
allows the opening of files that are beyond what the OS can safely handle.
The 32bit builds (including windows binaries) are implicitly data safe,
due to using the 2^31 file APIs.
The 64bit builds allow files from 2^31 (2^32 on RO5) to 2^63 to be opened
when they shouldn't.
This is the effect your patch has:
32bit builds 64bit builds
RO 3,4,6 Data Unsafe (no filesize limit) Data Unsafe (no filesize limit)
RO 5 Data Unsafe (no filesize limit) Data Unsafe (no filesize limit)
Your patch has the effect of increasing the cases in which user's data is
not protected. As such, we can not commit it.
I propose that we do the following instead, add a check in HostFS's file
open to prevent large files being opened (>2GB) and return a "file too
large" error to RISC OS, this meets the requirements of protecting the
users data in all scenarios.
The effects of this proposal are in this table:
32bit builds 64bit builds
RO 3,4,6 Data Safe (2GB filesize limit) Data Safe (2GB filesize limit)
RO 5 Data Safe (2GB filesize limit) Data Safe (2GB filesize limit)
A limitation of 2GB files on a RISC OS that supports 4GB files is not
ideal, but safe from data corruption. It is much safer than setting the
limit at 4GB and allowing 2-4GB files to be opened on a RISC OS that only
supports 2GB files.
If there is a backwards compatible way that a RISC OS file system module
can determine that it is running on an OS that supports files up to 4GB
in size, then this API could be used to modify HostFS to increase its
filesize limit to 4GB on supported RISC OSes.
This would need to be something much more robust than an OS version check,
and involve HostFS declaring it could use 4GB files as well as the OS
declaring it wants to use 4GB files (a negotiation).
Peter
--
Peter Howkins
peter.howkins@marutan.net
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
> In article <20160216173419.GA20045@spod.org>,
> > For the long explanation of why this HostFS patch is not included in
> > RPCEmu, please see this post from 2011.
> >
> > http://www.riscos.info/pipermail/rpcemu/2011-October/001383.html
>
> I recall that discussion at the time, and the thread continued
> http://www.riscos.info/pipermail/rpcemu/2011-October/001384.html
> then fell silent.
>
> The tests performed in the post you've highlighted didn't really elicit any
> new information:
> Old FileSwitch + 4GB files + vanilla HostFS => doesn't work
> Old FileSwitch + 4GB files + 4GB HostFS => doesn't work
> New FileSwitch + 4GB files + vanilla HostFS => doesn't work
> New FileSwitch + 4GB files + 4GB HostFS => does work
>
> the point being that having the underlying HostFS being 4GB capable is
> harmless for OS versions that it already doesn't work on, it's only when
> coupled with RISC OS 5.20 and (chronologically) later that you get benefit.
Unfortuanately it is not harmless on OSes < 5.20 nor harmless on >= 5.20.
At this time it seems to be important to state the requirements that
HostFS should have. These may only have been implicit before.
1) HostFS needs to work on all versions of RISC OS that RPCEmu can run (at
least 3.5, possibly even 3.1 for arcem support).
2) RPCEmu users can place any size file in their HostFS directory on the
Host Side, and HostFS must handle this gracefully.
3) HostFS should protect RPCEmu user's data from loss or corruption.
One specific case discussed here of 3) is that allowing files to be opened
that are larger than the maximum size that RO supports risks data loss, as
a program can only work on only some of the data in a file, which can cause
corruption.
Here follows a table describing the current situation in RPCEmu.
I have used the phrase 'Data Safe' to represent that they don't allow
files to be opened that are larger than OS supports.
32bit builds 64bit builds
RO 3,4,6 Data Safe (2GB filesize limit) Data Unsafe (no filesize limit)
RO 5 Data Safe (2GB filesize limit) Data Unsafe (no filesize limit)
The problem of RPCEmu as it stands at the moment is that it does not meet
our requirements of protecting users data on 64bit builds. Because HostFS
allows the opening of files that are beyond what the OS can safely handle.
The 32bit builds (including windows binaries) are implicitly data safe,
due to using the 2^31 file APIs.
The 64bit builds allow files from 2^31 (2^32 on RO5) to 2^63 to be opened
when they shouldn't.
This is the effect your patch has:
32bit builds 64bit builds
RO 3,4,6 Data Unsafe (no filesize limit) Data Unsafe (no filesize limit)
RO 5 Data Unsafe (no filesize limit) Data Unsafe (no filesize limit)
Your patch has the effect of increasing the cases in which user's data is
not protected. As such, we can not commit it.
I propose that we do the following instead, add a check in HostFS's file
open to prevent large files being opened (>2GB) and return a "file too
large" error to RISC OS, this meets the requirements of protecting the
users data in all scenarios.
The effects of this proposal are in this table:
32bit builds 64bit builds
RO 3,4,6 Data Safe (2GB filesize limit) Data Safe (2GB filesize limit)
RO 5 Data Safe (2GB filesize limit) Data Safe (2GB filesize limit)
A limitation of 2GB files on a RISC OS that supports 4GB files is not
ideal, but safe from data corruption. It is much safer than setting the
limit at 4GB and allowing 2-4GB files to be opened on a RISC OS that only
supports 2GB files.
If there is a backwards compatible way that a RISC OS file system module
can determine that it is running on an OS that supports files up to 4GB
in size, then this API could be used to modify HostFS to increase its
filesize limit to 4GB on supported RISC OSes.
This would need to be something much more robust than an OS version check,
and involve HostFS declaring it could use 4GB files as well as the OS
declaring it wants to use 4GB files (a negotiation).
Peter
--
Peter Howkins
peter.howkins@marutan.net
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Javascript features list
Is there a maintained list of available JS features ?
I found this http://wiki.netsurf-browser.org/Development_Plan
but it looks a bit out of date and I found some circular references
on http://www.netsurf-browser.org/documentation/info.html#JavaScript
I am looking at a survey form and wondered if buttons should work ?
The code in question is:
<input class="btn btn-large surveyFooterSubmitBtn"
id="SurveySubmitButtonElement"
onClick="if (this.form.validate)
{if (this.form.validate())
{this.disabled=true;}
else
{return false;}
}
postExecuteJavaScriptLogic();
customValidateAndSubmit(document.forms['run']);
this.disabled=true;
return false;"
TYPE="button" NAME="submitButton" VALUE="Continue">
I found this http://wiki.netsurf-browser.org/Development_Plan
but it looks a bit out of date and I found some circular references
on http://www.netsurf-browser.org/documentation/info.html#JavaScript
I am looking at a survey form and wondered if buttons should work ?
The code in question is:
<input class="btn btn-large surveyFooterSubmitBtn"
id="SurveySubmitButtonElement"
onClick="if (this.form.validate)
{if (this.form.validate())
{this.disabled=true;}
else
{return false;}
}
postExecuteJavaScriptLogic();
customValidateAndSubmit(document.forms['run']);
this.disabled=true;
return false;"
TYPE="button" NAME="submitButton" VALUE="Continue">
Re: Updating !Boot and !System
In article <1016645455.DaveMeUK@my.inbox.com>,
Dave Higton <dave@davehigton.me.uk> wrote:
> In message <55545d0a67chris@chris-johnson.org.uk>
> cj <chris@chris-johnson.org.uk> wrote:
> >In article <8206585455.DaveMeUK@my.inbox.com>,
> > Dave Higton <dave@davehigton.me.uk> wrote:
> >> What is this "installer module"?
> >
> >*h. installer
> >==> Help on keyword Installer
> >Module is: Installer 0.13 (19 Jul 2012)
>
> It still looks like a struggle to use it, though. According to
> the help, it can only cope with a single file per command, and
> it has to be given both the source path (which we know) and the
> destination path.
How many files are present in the !Boot/!System updates from NetSurf?
CI 3419 has
13 in !Cache
5 in !System
78 in !Unicode
so it's not beyond the wit of man to have 91 lines in an obey file, assuming
you use some tool to get their leaf names (or, I suppose, a bit of perl or
whatever to generate the script by the build system).
> And it still copies based on the file date (which we know we
> can't trust) and not the file content.
Not so. See section 5.2 of the spec
http://www.marutan.net/wikiref/Acorn%20Registered%20Developer%20REFERNC/RO4/API/HTML/CONFIGUR.HTM
> Does anyone have any examples of use of the installer module
> that I can look at, please?
Both the "PlingSystem" archive and 5.22 disc image from ROOL include obey
files using it. You can also programmatically start !Configure and open the
respective plugin, but that might be a bit spooky to watch,
Sprow.
Dave Higton <dave@davehigton.me.uk> wrote:
> In message <55545d0a67chris@chris-johnson.org.uk>
> cj <chris@chris-johnson.org.uk> wrote:
> >In article <8206585455.DaveMeUK@my.inbox.com>,
> > Dave Higton <dave@davehigton.me.uk> wrote:
> >> What is this "installer module"?
> >
> >*h. installer
> >==> Help on keyword Installer
> >Module is: Installer 0.13 (19 Jul 2012)
>
> It still looks like a struggle to use it, though. According to
> the help, it can only cope with a single file per command, and
> it has to be given both the source path (which we know) and the
> destination path.
How many files are present in the !Boot/!System updates from NetSurf?
CI 3419 has
13 in !Cache
5 in !System
78 in !Unicode
so it's not beyond the wit of man to have 91 lines in an obey file, assuming
you use some tool to get their leaf names (or, I suppose, a bit of perl or
whatever to generate the script by the build system).
> And it still copies based on the file date (which we know we
> can't trust) and not the file content.
Not so. See section 5.2 of the spec
http://www.marutan.net/wikiref/Acorn%20Registered%20Developer%20REFERNC/RO4/API/HTML/CONFIGUR.HTM
> Does anyone have any examples of use of the installer module
> that I can look at, please?
Both the "PlingSystem" archive and 5.22 disc image from ROOL include obey
files using it. You can also programmatically start !Configure and open the
respective plugin, but that might be a bit spooky to watch,
Sprow.
Re: netsurf-website: branch master updated. fe3dcc639b6d1f39f1b1e6d4846ff6ad1123d0d9
On Thu, 18 Feb 2016 20:56:02 +0000, Michael Drake wrote:
> On 18/02/16 19:22, Chris Young wrote:
>
> > Despite the filename, the netsurf-3.4.lha archive actually contains
> > v3.3 for some reason!
>
> Odd! For now I've just fixed the link to say 3.3.
>
> Vince is there a 3.4 for Amiga? Also, what about Atari and Haiku?
Vince, when you do build 3.4 for Amiga, please can you add commit
1828079518fbb4d4edf2daec47a7bc74027ff0c9
first. Something I changed recently stopped NetSurf from re-launching
if it failed to quit, unfortunately there was another long-standing
bug which was preventing it from quitting properly. I've only just
been alerted to it, but I consider this fix critical for the 3.4
release as it occurs in the default config.
It doesn't affect other platforms.
Thanks
Chris
> On 18/02/16 19:22, Chris Young wrote:
>
> > Despite the filename, the netsurf-3.4.lha archive actually contains
> > v3.3 for some reason!
>
> Odd! For now I've just fixed the link to say 3.3.
>
> Vince is there a 3.4 for Amiga? Also, what about Atari and Haiku?
Vince, when you do build 3.4 for Amiga, please can you add commit
1828079518fbb4d4edf2daec47a7bc74027ff0c9
first. Something I changed recently stopped NetSurf from re-launching
if it failed to quit, unfortunately there was another long-standing
bug which was preventing it from quitting properly. I've only just
been alerted to it, but I consider this fix critical for the 3.4
release as it occurs in the default config.
It doesn't affect other platforms.
Thanks
Chris
Re: Updating !Boot and !System
In article <1016645455.DaveMeUK@my.inbox.com>,
Dave Higton <dave@davehigton.me.uk> wrote:
> Does anyone have any examples of use of the installer module
> that I can look at, please?
It is used by BootMerge and SysMerge (in eg
Boot:RO500Hook.Res.Configure.!BootSetup), but those apps are not a
lot of help for usage. I guess the ROOL source is the place to look.
Rather OT for this list.
--
Chris Johnson
Dave Higton <dave@davehigton.me.uk> wrote:
> Does anyone have any examples of use of the installer module
> that I can look at, please?
It is used by BootMerge and SysMerge (in eg
Boot:RO500Hook.Res.Configure.!BootSetup), but those apps are not a
lot of help for usage. I guess the ROOL source is the place to look.
Rather OT for this list.
--
Chris Johnson
Re: [Rpcemu] RPCEmu Mac OS X test build of 0.8.14
Tim Powys-Lybbe, on 19 Feb, wrote:
>
> > On 15 Feb 2016, at 11:55 pm, Theo Markettos <theo@markettos.org.uk>
> > wrote:
> >
> > I've made a test build of 0.8.14 for Mac, which you can find at:
> > https://bitbucket.org/caliston/rpcemu-spoon-caliston/downloads
[snip]
> This time (I failed with 0.8.13) I read the small print very closely and
> have managed to get RISC running in a microscopic window of about 2 by 1.5
> inches.
Retina??
If so then ‘Get Info’ on RPCEmu.app and tick “Open in Low Resolutionâ€.
> I had to delete the RPCEmu.plist file and then reboot to get this far.
>
> However this micro-window read as:
>
> ARM 610 Processor
>
> Acorn ADFS
>
> Error: Use *Desktop to start TaskManager (Error number &81F403)
>
> So I typed in <* Desktop> which produced a micro desktop with this
> message:
>
> Machine startup not completed successfully: 'Disc drive not known'
*co. FileSystem HostFS
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
>
> > On 15 Feb 2016, at 11:55 pm, Theo Markettos <theo@markettos.org.uk>
> > wrote:
> >
> > I've made a test build of 0.8.14 for Mac, which you can find at:
> > https://bitbucket.org/caliston/rpcemu-spoon-caliston/downloads
[snip]
> This time (I failed with 0.8.13) I read the small print very closely and
> have managed to get RISC running in a microscopic window of about 2 by 1.5
> inches.
Retina??
If so then ‘Get Info’ on RPCEmu.app and tick “Open in Low Resolutionâ€.
> I had to delete the RPCEmu.plist file and then reboot to get this far.
>
> However this micro-window read as:
>
> ARM 610 Processor
>
> Acorn ADFS
>
> Error: Use *Desktop to start TaskManager (Error number &81F403)
>
> So I typed in <* Desktop> which produced a micro desktop with this
> message:
>
> Machine startup not completed successfully: 'Disc drive not known'
*co. FileSystem HostFS
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: [Rpcemu] RPCEmu Mac OS X test build of 0.8.14
> On 15 Feb 2016, at 11:55 pm, Theo Markettos <theo@markettos.org.uk> wrote:
>
> I've made a test build of 0.8.14 for Mac, which you can find at:
> https://bitbucket.org/caliston/rpcemu-spoon-caliston/downloads
>
> I've also applied a patch from Sprow to support 4GB files.
>
> I have only done the minimal amount of testing, but it seems to work for me.
> Other than rebuilding, outstanding problems (eg lack of Retina support) will
> remain.
>
> Please let me know if there are any new bugs this has introduced.
>
>
Many thanks for the continued Mac support.
This time (I failed with 0.8.13) I read the small print very closely and have managed to get RISC running in a microscopic window of about 2 by 1.5 inches.
I had to delete the RPCEmu.plist file and then reboot to get this far.
However this micro-window read as:
ARM 610 Processor
Acorn ADFS
Error: Use *Desktop to start TaskManager (Error number &81F403)
So I typed in <* Desktop> which produced a micro desktop with this message:
Machine startup not completed successfully: 'Disc drive not known'
I had put a copy of the latest HardDisc4 utils file, downloaded from the ROOL site, in the hostfs directory and it remains there, unexpanded and with no visible !Boot, etc.
The MIPS look rather impressive at an AVG of 1000 odd.
Any help, anyone?
--
Tim Powys-Lybbe tim@powys.org
for a miscellany of bygones: http://powys.org/
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
>
> I've made a test build of 0.8.14 for Mac, which you can find at:
> https://bitbucket.org/caliston/rpcemu-spoon-caliston/downloads
>
> I've also applied a patch from Sprow to support 4GB files.
>
> I have only done the minimal amount of testing, but it seems to work for me.
> Other than rebuilding, outstanding problems (eg lack of Retina support) will
> remain.
>
> Please let me know if there are any new bugs this has introduced.
>
>
Many thanks for the continued Mac support.
This time (I failed with 0.8.13) I read the small print very closely and have managed to get RISC running in a microscopic window of about 2 by 1.5 inches.
I had to delete the RPCEmu.plist file and then reboot to get this far.
However this micro-window read as:
ARM 610 Processor
Acorn ADFS
Error: Use *Desktop to start TaskManager (Error number &81F403)
So I typed in <* Desktop> which produced a micro desktop with this message:
Machine startup not completed successfully: 'Disc drive not known'
I had put a copy of the latest HardDisc4 utils file, downloaded from the ROOL site, in the hostfs directory and it remains there, unexpanded and with no visible !Boot, etc.
The MIPS look rather impressive at an AVG of 1000 odd.
Any help, anyone?
--
Tim Powys-Lybbe tim@powys.org
for a miscellany of bygones: http://powys.org/
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: Updating !Boot and !System
In message <55545d0a67chris@chris-johnson.org.uk>
cj <chris@chris-johnson.org.uk> wrote:
>In article <8206585455.DaveMeUK@my.inbox.com>,
> Dave Higton <dave@davehigton.me.uk> wrote:
>> What is this "installer module"?
>
>*h. installer
>==> Help on keyword Installer
>Module is: Installer 0.13 (19 Jul 2012)
[snip]
Thanks, Chris. I was unaware of the existence of the module!
It still looks like a struggle to use it, though. According to
the help, it can only cope with a single file per command, and
it has to be given both the source path (which we know) and the
destination path. So it still means iterating through all the
contents of each new partial !Boot and !System folders in the
NS archive, then inventing the corresponding destination path.
And it still copies based on the file date (which we know we
can't trust) and not the file content. OK, not such a big deal,
just more disc write operations than we would have in an ideal
world. Exactly as many as my proposed Obey file would cause,
as far as I can see.
There are references to a backup, which might be a handy safety
belt - although it's not clear to me where that backup might be
found.
Does anyone have any examples of use of the installer module
that I can look at, please?
Dave
____________________________________________________________
Can't remember your password? Do you need a strong and secure password?
Use Password manager! It stores your passwords & protects your account.
Check it out at http://mysecurelogon.com/password-manager
cj <chris@chris-johnson.org.uk> wrote:
>In article <8206585455.DaveMeUK@my.inbox.com>,
> Dave Higton <dave@davehigton.me.uk> wrote:
>> What is this "installer module"?
>
>*h. installer
>==> Help on keyword Installer
>Module is: Installer 0.13 (19 Jul 2012)
[snip]
Thanks, Chris. I was unaware of the existence of the module!
It still looks like a struggle to use it, though. According to
the help, it can only cope with a single file per command, and
it has to be given both the source path (which we know) and the
destination path. So it still means iterating through all the
contents of each new partial !Boot and !System folders in the
NS archive, then inventing the corresponding destination path.
And it still copies based on the file date (which we know we
can't trust) and not the file content. OK, not such a big deal,
just more disc write operations than we would have in an ideal
world. Exactly as many as my proposed Obey file would cause,
as far as I can see.
There are references to a backup, which might be a handy safety
belt - although it's not clear to me where that backup might be
found.
Does anyone have any examples of use of the installer module
that I can look at, please?
Dave
____________________________________________________________
Can't remember your password? Do you need a strong and secure password?
Use Password manager! It stores your passwords & protects your account.
Check it out at http://mysecurelogon.com/password-manager
Re: Updating !Boot and !System
In article <8206585455.DaveMeUK@my.inbox.com>,
Dave Higton <dave@davehigton.me.uk> wrote:
> What is this "installer module"?
*h. installer
==> Help on keyword Installer
Module is: Installer 0.13 (19 Jul 2012)
Commands provided:
Install_LogDir Install_NoLogDir Install_Update Install_Remove
Install_Merge Install_DeMerge Install_CheckSystem
Pulled out of the module:
*Install_CheckSystem ensures that versions of modules present in
OS-version subdirectories of !System are not older than those in
previous-version subdirectories, or the root Modules subdirectory.
Syntax: *Install_CheckSystem
*Install_DeMerge strips entries from a structured Obey, Desktop or
TaskObey file. The changes file must include at least the entry
headers for the most recent version of the entry to be removed.
Syntax: *Install_DeMerge <base file> <changes file> <destination file>
*Install_Merge combines the contents of two structured Obey, Desktop
or TaskObey files, using the most up-to-date entries from each.
Ordering of entries is determined from the base file, or failing
that, from the rules in the changes file.
Syntax: *Install_Merge <base file> <changes file> <destination file>
*Install_Remove backs up and removes a file or directory.
Syntax: *Install_Remove <file spec
*Install_Update installs a file, subject to appropriate newness
criteria.
Syntax: *Install_Update <source file> <destination file>
*Install_NoLogDir forgets the location of the log file and backup
directory.
Syntax: *Install_NoLogDir
*Install_LogDir sets a new directory in which the log file will be
written and backups will be stored. If no directory is stated, the
current log directory is printed.
Syntax: *Install_LogDir [<new log direct
--
Chris Johnson
Dave Higton <dave@davehigton.me.uk> wrote:
> What is this "installer module"?
*h. installer
==> Help on keyword Installer
Module is: Installer 0.13 (19 Jul 2012)
Commands provided:
Install_LogDir Install_NoLogDir Install_Update Install_Remove
Install_Merge Install_DeMerge Install_CheckSystem
Pulled out of the module:
*Install_CheckSystem ensures that versions of modules present in
OS-version subdirectories of !System are not older than those in
previous-version subdirectories, or the root Modules subdirectory.
Syntax: *Install_CheckSystem
*Install_DeMerge strips entries from a structured Obey, Desktop or
TaskObey file. The changes file must include at least the entry
headers for the most recent version of the entry to be removed.
Syntax: *Install_DeMerge <base file> <changes file> <destination file>
*Install_Merge combines the contents of two structured Obey, Desktop
or TaskObey files, using the most up-to-date entries from each.
Ordering of entries is determined from the base file, or failing
that, from the rules in the changes file.
Syntax: *Install_Merge <base file> <changes file> <destination file>
*Install_Remove backs up and removes a file or directory.
Syntax: *Install_Remove <file spec
*Install_Update installs a file, subject to appropriate newness
criteria.
Syntax: *Install_Update <source file> <destination file>
*Install_NoLogDir forgets the location of the log file and backup
directory.
Syntax: *Install_NoLogDir
*Install_LogDir sets a new directory in which the log file will be
written and backups will be stored. If no directory is stated, the
current log directory is printed.
Syntax: *Install_LogDir [<new log direct
--
Chris Johnson
Re: Updating !Boot and !System
In message <55540fa3b0webpages@sprow.co.uk>
Sprow <webpages@sprow.co.uk> wrote:
>In article <11a00d5455.DaveMeUK@my.inbox.com>,
> Dave Higton <dave@davehigton.me.uk> wrote:
>> I'm proposing that the attached Obey file be included in NetSurf
>> zips for RISC OS. If double-clicked, it should perform the same
>> update of !Boot and !System as if done manually through the
>> Configure application.
>
>That's a bit crude, and nothing like the process !Configure uses.
>
>Use the installer module - it's what it's there for, and gives you all the
>side effects as though done properly with Boot Merge and Sys Merge, like
>backing up older modules and recording any updates in the log,
What is this "installer module"?
Dave
____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
Check it out at http://www.inbox.com/earth
Sprow <webpages@sprow.co.uk> wrote:
>In article <11a00d5455.DaveMeUK@my.inbox.com>,
> Dave Higton <dave@davehigton.me.uk> wrote:
>> I'm proposing that the attached Obey file be included in NetSurf
>> zips for RISC OS. If double-clicked, it should perform the same
>> update of !Boot and !System as if done manually through the
>> Configure application.
>
>That's a bit crude, and nothing like the process !Configure uses.
>
>Use the installer module - it's what it's there for, and gives you all the
>side effects as though done properly with Boot Merge and Sys Merge, like
>backing up older modules and recording any updates in the log,
What is this "installer module"?
Dave
____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
Check it out at http://www.inbox.com/earth
Thursday, 18 February 2016
Re: Updating !Boot and !System
In article <11a00d5455.DaveMeUK@my.inbox.com>,
Dave Higton <dave@davehigton.me.uk> wrote:
> I'm proposing that the attached Obey file be included in NetSurf
> zips for RISC OS. If double-clicked, it should perform the same
> update of !Boot and !System as if done manually through the
> Configure application.
That's a bit crude, and nothing like the process !Configure uses.
Use the installer module - it's what it's there for, and gives you all the
side effects as though done properly with Boot Merge and Sys Merge, like
backing up older modules and recording any updates in the log,
Sprow.
Dave Higton <dave@davehigton.me.uk> wrote:
> I'm proposing that the attached Obey file be included in NetSurf
> zips for RISC OS. If double-clicked, it should perform the same
> update of !Boot and !System as if done manually through the
> Configure application.
That's a bit crude, and nothing like the process !Configure uses.
Use the installer module - it's what it's there for, and gives you all the
side effects as though done properly with Boot Merge and Sys Merge, like
backing up older modules and recording any updates in the log,
Sprow.
Updating !Boot and !System
I'm proposing that the attached Obey file be included in NetSurf
zips for RISC OS. If double-clicked, it should perform the same
update of !Boot and !System as if done manually through the
Configure application.
This follows yesterday evening's dicussion on IRC, which seemed
to peter out about 22:10.
Dave
____________________________________________________________
Can't remember your password? Do you need a strong and secure password?
Use Password manager! It stores your passwords & protects your account.
Check it out at http://mysecurelogon.com/manager
zips for RISC OS. If double-clicked, it should perform the same
update of !Boot and !System as if done manually through the
Configure application.
This follows yesterday evening's dicussion on IRC, which seemed
to peter out about 22:10.
Dave
____________________________________________________________
Can't remember your password? Do you need a strong and secure password?
Use Password manager! It stores your passwords & protects your account.
Check it out at http://mysecurelogon.com/manager
Netsurf frame buffer - input device
Hello
We are trying to run the netsurf frame buffer, in order to use input devices.
>From your site it seems that the "linux" option (framebuffer) enables the input devices. (from /dev/input by default)
On the other hand, when we look at the code in libnsfb, surface/linux.c, the linux_input is a stub function.
Does netsurf support input device when working with frame buffer?
Thanks,
Michal
Re: netsurf-website: branch master updated. fe3dcc639b6d1f39f1b1e6d4846ff6ad1123d0d9
On 18/02/16 19:22, Chris Young wrote:
> Despite the filename, the netsurf-3.4.lha archive actually contains
> v3.3 for some reason!
Odd! For now I've just fixed the link to say 3.3.
Vince is there a 3.4 for Amiga? Also, what about Atari and Haiku?
Cheers,
--
Michael Drake http://www.netsurf-browser.org/
> Despite the filename, the netsurf-3.4.lha archive actually contains
> v3.3 for some reason!
Odd! For now I've just fixed the link to say 3.3.
Vince is there a 3.4 for Amiga? Also, what about Atari and Haiku?
Cheers,
--
Michael Drake http://www.netsurf-browser.org/
Re: netsurf-website: branch master updated. fe3dcc639b6d1f39f1b1e6d4846ff6ad1123d0d9
On Thu, 18 Feb 2016 18:21:46 +0000, Commit Mailer wibbled on for an age:
> -<p class="downloadmain downloadfirst"><a href="http://download.netsurf-browser.org/netsurf/releases/pre-built/amiga/netsurf-3.2.lha"><span>NetSurf 3.2 for AmigaOS</span> <span>(5.9MB)</span> <span class="downloaddate">30 Aug 2014</span></a></p>
> +<p class="downloadmain downloadfirst"><a href="http://download.netsurf-browser.org/netsurf/releases/pre-built/amiga/netsurf-3.4.lha"><span>NetSurf 3.4 for AmigaOS</span> <span>(5.9MB)</span> <span class="downloaddate">16 Feb 2016</span></a></p>
Despite the filename, the netsurf-3.4.lha archive actually contains
v3.3 for some reason!
Chris
> -<p class="downloadmain downloadfirst"><a href="http://download.netsurf-browser.org/netsurf/releases/pre-built/amiga/netsurf-3.2.lha"><span>NetSurf 3.2 for AmigaOS</span> <span>(5.9MB)</span> <span class="downloaddate">30 Aug 2014</span></a></p>
> +<p class="downloadmain downloadfirst"><a href="http://download.netsurf-browser.org/netsurf/releases/pre-built/amiga/netsurf-3.4.lha"><span>NetSurf 3.4 for AmigaOS</span> <span>(5.9MB)</span> <span class="downloaddate">16 Feb 2016</span></a></p>
Despite the filename, the netsurf-3.4.lha archive actually contains
v3.3 for some reason!
Chris
Wednesday, 17 February 2016
Re: JavaScript status indicator - request for a snippet of code
In article <b067675355.ricp@user.minijem.plus.com>,
Richard Porter <ricp@minijem.plus.com> wrote:
> > <script> document.write("javascript is enabled"); </script>
> alternatively / additionally:
> <noscript>JS Off</noscript>
Excellent!
I drop out of PHP and do them both, then drop back in:
?>
<script> document.write("<font color=red>JavaScript is
enabled!</font><br>"); </script>
<noscript>JavaScript is disabled<br></noscript>
<?php
Thank you both!
John
--
| John Williams
| johnrw@ukgateway.net
I am NOT argumentative! *
Richard Porter <ricp@minijem.plus.com> wrote:
> > <script> document.write("javascript is enabled"); </script>
> alternatively / additionally:
> <noscript>JS Off</noscript>
Excellent!
I drop out of PHP and do them both, then drop back in:
?>
<script> document.write("<font color=red>JavaScript is
enabled!</font><br>"); </script>
<noscript>JavaScript is disabled<br></noscript>
<?php
Thank you both!
John
--
| John Williams
| johnrw@ukgateway.net
I am NOT argumentative! *
Re: JavaScript status indicator - request for a snippet of code
On 17 Feb 2016 John Rickman Iyonix wrote:
>> I suspect that what I require is very simple/trivial and can be done in one
>> line of JS.
> try this:
> <script> document.write("javascript is enabled"); </script>
alternatively / additionally:
<noscript>JS Off</noscript>
--
Richard Porter http://www.minijem.plus.com/
Skype: minijem2 mailto:ricp@minijem.plus.com
I don't want a "user experience" - I just want stuff that works.
>> I suspect that what I require is very simple/trivial and can be done in one
>> line of JS.
> try this:
> <script> document.write("javascript is enabled"); </script>
alternatively / additionally:
<noscript>JS Off</noscript>
--
Richard Porter http://www.minijem.plus.com/
Skype: minijem2 mailto:ricp@minijem.plus.com
I don't want a "user experience" - I just want stuff that works.
Re: JavaScript status indicator - request for a snippet of code
In article <60ac4d5355.iyojohn@rickman.argonet.co.uk>,
John Rickman Iyonix <rickman@argonet.co.uk> wrote:
> try this:
> <script> document.write("javascript is enabled"); </script>
Perfick, as HE Bates was prone to write!
Thank you.
John
--
| John Williams
| johnrw@ukgateway.net
Names for Soul Band:- Soul Suppliers *
John Rickman Iyonix <rickman@argonet.co.uk> wrote:
> try this:
> <script> document.write("javascript is enabled"); </script>
Perfick, as HE Bates was prone to write!
Thank you.
John
--
| John Williams
| johnrw@ukgateway.net
Names for Soul Band:- Soul Suppliers *
Re: JavaScript status indicator - request for a snippet of code
John Williams wrote
> Can anyone who understands JS provide a snippet of JS code that I can
> insert in the footer of my local HTML index page to conditionally print,
> say, "JavaScript is enabled!" as a reminder of NetSurf's current JS status.
> If anyone wants to "go the whole hog" and reflect the two possible states,
> then my local index page¹ uses PHP served by WebJames, but that is by no
> means essential and may well be over-complicated.
> I suspect that what I require is very simple/trivial and can be done in one
> line of JS.
try this:
<script> document.write("javascript is enabled"); </script>
John
--
John Rickman - http://rickman.orpheusweb.co.uk/lynx
Doth not even nature itself teach you, that, if a man have long hair,
it is a shame unto him? Saint Paul KJ
> Can anyone who understands JS provide a snippet of JS code that I can
> insert in the footer of my local HTML index page to conditionally print,
> say, "JavaScript is enabled!" as a reminder of NetSurf's current JS status.
> If anyone wants to "go the whole hog" and reflect the two possible states,
> then my local index page¹ uses PHP served by WebJames, but that is by no
> means essential and may well be over-complicated.
> I suspect that what I require is very simple/trivial and can be done in one
> line of JS.
try this:
<script> document.write("javascript is enabled"); </script>
John
--
John Rickman - http://rickman.orpheusweb.co.uk/lynx
Doth not even nature itself teach you, that, if a man have long hair,
it is a shame unto him? Saint Paul KJ
JavaScript status indicator - request for a snippet of code
Can anyone who understands JS provide a snippet of JS code that I can
insert in the footer of my local HTML index page to conditionally print,
say, "JavaScript is enabled!" as a reminder of NetSurf's current JS status.
If anyone wants to "go the whole hog" and reflect the two possible states,
then my local index page¹ uses PHP served by WebJames, but that is by no
means essential and may well be over-complicated.
I suspect that what I require is very simple/trivial and can be done in one
line of JS.
TIA
John
¹ As described in my article in Archive magazine some years ago.
--
| John Williams
| johnrw@ukgateway.net
Names for Soul Band:- Soul Doubt *
insert in the footer of my local HTML index page to conditionally print,
say, "JavaScript is enabled!" as a reminder of NetSurf's current JS status.
If anyone wants to "go the whole hog" and reflect the two possible states,
then my local index page¹ uses PHP served by WebJames, but that is by no
means essential and may well be over-complicated.
I suspect that what I require is very simple/trivial and can be done in one
line of JS.
TIA
John
¹ As described in my article in Archive magazine some years ago.
--
| John Williams
| johnrw@ukgateway.net
Names for Soul Band:- Soul Doubt *
Re: [Rpcemu] RpcEmu installation issue
David Pitt, on 17 Feb, wrote:
> John Rickman Iyonix, on 16 Feb, wrote:
>
> > George Buttigieg wrote
> >
> > > Hi, I've been following the instructions at
> > > http://www.marutan.net/rpcemu/manual/#install, to install rpcemu (v.
> > > 0.8.14) onto my Linux desktop (Ubuntu 14.04). These instructions
> > > worked fine, and I managed to get RiscOs 3.71 running fine.
> >
> > > I then followed a link to
> > > http://www.4corn.co.uk/articles/rpcemu371win/ , which suggested
> > > installing a 'Universal Boot' onto the hostfs. Although these
> > > instructions came under the heading of a Windows installation, I
> > > assumed they would also be relevant to a Linux install. However, I now
> > > get an error each time I reboot. It's difficult to read the error logs
> > > in the console window, but when the desktop appears, I get an alert,
> > > and when I click the 'Describe' button, it says 'SWI &A0F00E not
> > > known'.
> >
> > I have the same problem on a windows 10 machine. Ie 'SWI &A0F00E not
> > known'. If I click it out the system seems to be ok.
>
> Same here on both the Mac and Windows 7.
>
> It goes away if VRAM is set to zero here!! Hmm!!
P.S. 0.8.13 doesn't error.
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
> John Rickman Iyonix, on 16 Feb, wrote:
>
> > George Buttigieg wrote
> >
> > > Hi, I've been following the instructions at
> > > http://www.marutan.net/rpcemu/manual/#install, to install rpcemu (v.
> > > 0.8.14) onto my Linux desktop (Ubuntu 14.04). These instructions
> > > worked fine, and I managed to get RiscOs 3.71 running fine.
> >
> > > I then followed a link to
> > > http://www.4corn.co.uk/articles/rpcemu371win/ , which suggested
> > > installing a 'Universal Boot' onto the hostfs. Although these
> > > instructions came under the heading of a Windows installation, I
> > > assumed they would also be relevant to a Linux install. However, I now
> > > get an error each time I reboot. It's difficult to read the error logs
> > > in the console window, but when the desktop appears, I get an alert,
> > > and when I click the 'Describe' button, it says 'SWI &A0F00E not
> > > known'.
> >
> > I have the same problem on a windows 10 machine. Ie 'SWI &A0F00E not
> > known'. If I click it out the system seems to be ok.
>
> Same here on both the Mac and Windows 7.
>
> It goes away if VRAM is set to zero here!! Hmm!!
P.S. 0.8.13 doesn't error.
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Tuesday, 16 February 2016
Re: [Rpcemu] RpcEmu installation issue
John Rickman Iyonix, on 16 Feb, wrote:
> George Buttigieg wrote
>
> > Hi, I've been following the instructions at
> > http://www.marutan.net/rpcemu/manual/#install, to install rpcemu (v.
> > 0.8.14) onto my Linux desktop (Ubuntu 14.04). These instructions worked
> > fine, and I managed to get RiscOs 3.71 running fine.
>
> > I then followed a link to http://www.4corn.co.uk/articles/rpcemu371win/
> > , which suggested installing a 'Universal Boot' onto the hostfs.
> > Although these instructions came under the heading of a Windows
> > installation, I assumed they would also be relevant to a Linux install.
> > However, I now get an error each time I reboot. It's difficult to read
> > the error logs in the console window, but when the desktop appears, I
> > get an alert, and when I click the 'Describe' button, it says 'SWI
> > &A0F00E not known'.
>
> I have the same problem on a windows 10 machine. Ie 'SWI &A0F00E not
> known'. If I click it out the system seems to be ok.
Same here on both the Mac and Windows 7.
It goes away if VRAM is set to zero here!! Hmm!!
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
> George Buttigieg wrote
>
> > Hi, I've been following the instructions at
> > http://www.marutan.net/rpcemu/manual/#install, to install rpcemu (v.
> > 0.8.14) onto my Linux desktop (Ubuntu 14.04). These instructions worked
> > fine, and I managed to get RiscOs 3.71 running fine.
>
> > I then followed a link to http://www.4corn.co.uk/articles/rpcemu371win/
> > , which suggested installing a 'Universal Boot' onto the hostfs.
> > Although these instructions came under the heading of a Windows
> > installation, I assumed they would also be relevant to a Linux install.
> > However, I now get an error each time I reboot. It's difficult to read
> > the error logs in the console window, but when the desktop appears, I
> > get an alert, and when I click the 'Describe' button, it says 'SWI
> > &A0F00E not known'.
>
> I have the same problem on a windows 10 machine. Ie 'SWI &A0F00E not
> known'. If I click it out the system seems to be ok.
Same here on both the Mac and Windows 7.
It goes away if VRAM is set to zero here!! Hmm!!
--
David Pitt
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: [Rpcemu] RpcEmu installation issue
George Buttigieg wrote
> Hi, I've been following the instructions at
> http://www.marutan.net/rpcemu/manual/#install, to install rpcemu (v.
> 0.8.14) onto my Linux desktop (Ubuntu 14.04). These instructions
> worked fine, and I managed to get RiscOs 3.71 running fine.
> I then followed a link to
> http://www.4corn.co.uk/articles/rpcemu371win/ , which suggested
> installing a 'Universal Boot' onto the hostfs. Although these
> instructions came under the heading of a Windows installation, I
> assumed they would also be relevant to a Linux install. However, I now
> get an error each time I reboot. It's difficult to read the error logs
> in the console window, but when the desktop appears, I get an alert,
> and when I click the 'Describe' button, it says 'SWI &A0F00E not
> known'.
I have the same problem on a windows 10 machine. Ie 'SWI &A0F00E not
known'. If I click it out the system seems to be ok.
john
--
John Rickman - http://rickman.orpheusweb.co.uk/lynx
My sister Dolly says she won't boil the giblets. She says it makes the
pekinese too excited. DA
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
> Hi, I've been following the instructions at
> http://www.marutan.net/rpcemu/manual/#install, to install rpcemu (v.
> 0.8.14) onto my Linux desktop (Ubuntu 14.04). These instructions
> worked fine, and I managed to get RiscOs 3.71 running fine.
> I then followed a link to
> http://www.4corn.co.uk/articles/rpcemu371win/ , which suggested
> installing a 'Universal Boot' onto the hostfs. Although these
> instructions came under the heading of a Windows installation, I
> assumed they would also be relevant to a Linux install. However, I now
> get an error each time I reboot. It's difficult to read the error logs
> in the console window, but when the desktop appears, I get an alert,
> and when I click the 'Describe' button, it says 'SWI &A0F00E not
> known'.
I have the same problem on a windows 10 machine. Ie 'SWI &A0F00E not
known'. If I click it out the system seems to be ok.
john
--
John Rickman - http://rickman.orpheusweb.co.uk/lynx
My sister Dolly says she won't boil the giblets. She says it makes the
pekinese too excited. DA
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: [Rpcemu] RPCEmu 0.8.14
Bryan Hogan <rpcemu@helpful.demon.co.uk> wrote:
> In message <21d96d5055.Jan-Jaap@armx6.gmx.com>
> Jan-Jaap van der Geer <jjvdgeer@inbox.com> wrote:
> > How does this work? I run RISC OS 4.37 but trying to change to
> > a mode that would take more memory than 2MB the mode picker
> > automatically chooses less colours, so it does not seem to
> > support more than 2MB VRAM.
> > In the settings I have chosen "Risc PC - ARM610" and "VRAM 2Mb
> > (8MB if OS supported)". (RAM is 32Mb)
> > Anything else I should do?
> I would guess you need to increase the BandLimit value:
> <Boot$Dir>.Choices.Hardware.Disabled.Boot.PreDesk.Configure.!Run
> Otherwise RISC OS will think it can't get the data out of the
> memory quick enough, but that's irrelevant in an emulator so
> stick an extra zero on the end!
Good point. And it worked for 4.39 but not for 4.37. Not sure why I
had 4.37 anyway, so I'm on 4.39 now...
Well, I do know why I was on 4.37. I can't locate my installation
CD (or was it a download?) for 4.39. I did have the ROM image. I
was actually starting up on 4.39 and softloading 4.37 on it... But
I suppose the !Boot directory is still 4.37. Oh well. Not using it
too much anyway.
> Anyone in the London area can hear answers to this question and
> more at next Monday's ROUGOL meeting:
> http://www.rougol.jellybaby.net/meetings/index.html
> :-)
Sorry, a bit too far away from Oslo :)
Cheers,
Jan-Jaap
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
> In message <21d96d5055.Jan-Jaap@armx6.gmx.com>
> Jan-Jaap van der Geer <jjvdgeer@inbox.com> wrote:
> > How does this work? I run RISC OS 4.37 but trying to change to
> > a mode that would take more memory than 2MB the mode picker
> > automatically chooses less colours, so it does not seem to
> > support more than 2MB VRAM.
> > In the settings I have chosen "Risc PC - ARM610" and "VRAM 2Mb
> > (8MB if OS supported)". (RAM is 32Mb)
> > Anything else I should do?
> I would guess you need to increase the BandLimit value:
> <Boot$Dir>.Choices.Hardware.Disabled.Boot.PreDesk.Configure.!Run
> Otherwise RISC OS will think it can't get the data out of the
> memory quick enough, but that's irrelevant in an emulator so
> stick an extra zero on the end!
Good point. And it worked for 4.39 but not for 4.37. Not sure why I
had 4.37 anyway, so I'm on 4.39 now...
Well, I do know why I was on 4.37. I can't locate my installation
CD (or was it a download?) for 4.39. I did have the ROM image. I
was actually starting up on 4.39 and softloading 4.37 on it... But
I suppose the !Boot directory is still 4.37. Oh well. Not using it
too much anyway.
> Anyone in the London area can hear answers to this question and
> more at next Monday's ROUGOL meeting:
> http://www.rougol.jellybaby.net/meetings/index.html
> :-)
Sorry, a bit too far away from Oslo :)
Cheers,
Jan-Jaap
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
[Rpcemu] RpcEmu installation issue
Hi, I've been following the instructions at http://www.marutan.net/rpcemu/manual/#install, to install rpcemu (v. 0.8.14) onto my Linux desktop (Ubuntu 14.04). These instructions worked fine, and I managed to get RiscOs 3.71 running fine.
I then followed a link to http://www.4corn.co.uk/articles/rpcemu371win/ , which suggested installing a 'Universal Boot' onto the hostfs. Although these instructions came under the heading of a Windows installation, I assumed they would also be relevant to a Linux install. However, I now get an error each time I reboot. It's difficult to read the error logs in the console window, but when the desktop appears, I get an alert, and when I click the 'Describe' button, it says 'SWI &A0F00E not known'.
Should I not be attempting to boot from hostfs?
Another quite different question, regarding old archimedes games. I found a bunch of images available at http://www.emuparadise.me/Acorn_Archimedes_ROMs/58, and was wondering whether these would run under rpcemu? I gave one of the games a try (Lemmings), but simply got a black screen when I tried to start the app.
Many thanks for your sterling work,
George
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
I then followed a link to http://www.4corn.co.uk/articles/rpcemu371win/ , which suggested installing a 'Universal Boot' onto the hostfs. Although these instructions came under the heading of a Windows installation, I assumed they would also be relevant to a Linux install. However, I now get an error each time I reboot. It's difficult to read the error logs in the console window, but when the desktop appears, I get an alert, and when I click the 'Describe' button, it says 'SWI &A0F00E not known'.
Should I not be attempting to boot from hostfs?
Another quite different question, regarding old archimedes games. I found a bunch of images available at http://www.emuparadise.me/Acorn_Archimedes_ROMs/58, and was wondering whether these would run under rpcemu? I gave one of the games a try (Lemmings), but simply got a black screen when I tried to start the app.
Many thanks for your sterling work,
George
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
> Date: Sun, 14 Feb 2016 01:37:31 GMT
> From: Harriet Bazley <lists@orange.wingsandbeaks.org.uk>
> Subject: Re: Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
> On 12 Feb 2016 as I do recall,
> Michael Drake wrote:
>
> >
> > Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
> > ==============================================================
> >
> > Over the last couple of weeks I have been lucky enough to be able
> > to do work on the NetSurf project at work.
> > At [Codethink](http://www.codethink.co.uk/) we have an arrangement
> > that allows engineers to fill time between projects by making
> > contributions to certain open source projects.
> >
> > My request to work on [NetSurf](http://www.netsurf-browser.org/),
> > and more particularly, its [LibDOM](http://ns-b.org/projects/libdom/)
> > and [LibCSS](http://ns-b.org/projects/libcss/) sub-projects was
> > approved and I was able to get a good amount done.
>
> [snip]
>
> I'm impressed that so much of this seems to relate to making Netsurf faster
> and more efficient!
>
Michael,
Great work. I am so impressed that your company is forward thinking
enough to let people work on other projects that will deepen and
widen their skills.
More companies should have this level of faith in their developers.
Peter
> From: Harriet Bazley <lists@orange.wingsandbeaks.org.uk>
> Subject: Re: Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
> On 12 Feb 2016 as I do recall,
> Michael Drake wrote:
>
> >
> > Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
> > ==============================================================
> >
> > Over the last couple of weeks I have been lucky enough to be able
> > to do work on the NetSurf project at work.
> > At [Codethink](http://www.codethink.co.uk/) we have an arrangement
> > that allows engineers to fill time between projects by making
> > contributions to certain open source projects.
> >
> > My request to work on [NetSurf](http://www.netsurf-browser.org/),
> > and more particularly, its [LibDOM](http://ns-b.org/projects/libdom/)
> > and [LibCSS](http://ns-b.org/projects/libcss/) sub-projects was
> > approved and I was able to get a good amount done.
>
> [snip]
>
> I'm impressed that so much of this seems to relate to making Netsurf faster
> and more efficient!
>
Michael,
Great work. I am so impressed that your company is forward thinking
enough to let people work on other projects that will deepen and
widen their skills.
More companies should have this level of faith in their developers.
Peter
Re: [Rpcemu] RPCEmu Mac OS X test build of 0.8.14
In article <20160216173419.GA20045@spod.org>,
Peter Howkins <rpcemu.howkins@marutan.net> wrote:
> On Mon, Feb 15, 2016 at 11:55:44PM +0000, Theo Markettos wrote:
> >
> > I've also applied a patch from Sprow to support 4GB files.
> >
> > Please let me know if there are any new bugs this has introduced.
>
> For the long explanation of why this HostFS patch is not included in
> RPCEmu, please see this post from 2011.
>
> http://www.riscos.info/pipermail/rpcemu/2011-October/001383.html
I recall that discussion at the time, and the thread continued
http://www.riscos.info/pipermail/rpcemu/2011-October/001384.html
then fell silent.
The tests performed in the post you've highlighted didn't really elicit any
new information:
Old FileSwitch + 4GB files + vanilla HostFS => doesn't work
Old FileSwitch + 4GB files + 4GB HostFS => doesn't work
New FileSwitch + 4GB files + vanilla HostFS => doesn't work
New FileSwitch + 4GB files + 4GB HostFS => does work
the point being that having the underlying HostFS being 4GB capable is
harmless for OS versions that it already doesn't work on, it's only when
coupled with RISC OS 5.20 and (chronologically) later that you get benefit.
RISC OS 3/4/6 don't handle > 2G files gracefully out of the tin, and nor do
they with an updated HostFS.
Patch attached for 0.8.14 for reference,
Sprow.
Peter Howkins <rpcemu.howkins@marutan.net> wrote:
> On Mon, Feb 15, 2016 at 11:55:44PM +0000, Theo Markettos wrote:
> >
> > I've also applied a patch from Sprow to support 4GB files.
> >
> > Please let me know if there are any new bugs this has introduced.
>
> For the long explanation of why this HostFS patch is not included in
> RPCEmu, please see this post from 2011.
>
> http://www.riscos.info/pipermail/rpcemu/2011-October/001383.html
I recall that discussion at the time, and the thread continued
http://www.riscos.info/pipermail/rpcemu/2011-October/001384.html
then fell silent.
The tests performed in the post you've highlighted didn't really elicit any
new information:
Old FileSwitch + 4GB files + vanilla HostFS => doesn't work
Old FileSwitch + 4GB files + 4GB HostFS => doesn't work
New FileSwitch + 4GB files + vanilla HostFS => doesn't work
New FileSwitch + 4GB files + 4GB HostFS => does work
the point being that having the underlying HostFS being 4GB capable is
harmless for OS versions that it already doesn't work on, it's only when
coupled with RISC OS 5.20 and (chronologically) later that you get benefit.
RISC OS 3/4/6 don't handle > 2G files gracefully out of the tin, and nor do
they with an updated HostFS.
Patch attached for 0.8.14 for reference,
Sprow.
Re: [Rpcemu] RPCEmu Mac OS X test build of 0.8.14
On Mon, Feb 15, 2016 at 11:55:44PM +0000, Theo Markettos wrote:
>
> I've also applied a patch from Sprow to support 4GB files.
>
> Please let me know if there are any new bugs this has introduced.
For the long explanation of why this HostFS patch is not included in
RPCEmu, please see this post from 2011.
http://www.riscos.info/pipermail/rpcemu/2011-October/001383.html
Peter
--
Peter Howkins
peter.howkins@marutan.net
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
>
> I've also applied a patch from Sprow to support 4GB files.
>
> Please let me know if there are any new bugs this has introduced.
For the long explanation of why this HostFS patch is not included in
RPCEmu, please see this post from 2011.
http://www.riscos.info/pipermail/rpcemu/2011-October/001383.html
Peter
--
Peter Howkins
peter.howkins@marutan.net
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: Very slow download
On Tue, Feb 16, 2016 at 11:23:09AM +0000, Brian wrote:
> In article <20160216093631.GV7205@platypus.pepperfish.net>, Rob Kendrick
> <rjek@netsurf-browser.org> wrote:
> > On Tue, Feb 16, 2016 at 09:23:45AM +0000, Brian wrote:
> > > #3410 downloaded this morning, unusually, very slowly. Any reason,
> > > perhaps?
>
> > 2016-02-16 09:36:12 (3.64 MB/s) - NetSurf-gcc--3410.zip' saved
> > [3842257/3842257]
>
> > Works for me.
>
> #3413 (104kB/s) which works, sort of.
This kind of suggests the issue is at your end, or the ISP you use.
B.
> In article <20160216093631.GV7205@platypus.pepperfish.net>, Rob Kendrick
> <rjek@netsurf-browser.org> wrote:
> > On Tue, Feb 16, 2016 at 09:23:45AM +0000, Brian wrote:
> > > #3410 downloaded this morning, unusually, very slowly. Any reason,
> > > perhaps?
>
> > 2016-02-16 09:36:12 (3.64 MB/s) - NetSurf-gcc--3410.zip' saved
> > [3842257/3842257]
>
> > Works for me.
>
> #3413 (104kB/s) which works, sort of.
This kind of suggests the issue is at your end, or the ISP you use.
B.
Re: Very slow download
In article <20160216093631.GV7205@platypus.pepperfish.net>, Rob Kendrick
<rjek@netsurf-browser.org> wrote:
> On Tue, Feb 16, 2016 at 09:23:45AM +0000, Brian wrote:
> > #3410 downloaded this morning, unusually, very slowly. Any reason,
> > perhaps?
> 2016-02-16 09:36:12 (3.64 MB/s) - NetSurf-gcc--3410.zip' saved
> [3842257/3842257]
> Works for me.
> B.
#3413 (104kB/s) which works, sort of.
<rjek@netsurf-browser.org> wrote:
> On Tue, Feb 16, 2016 at 09:23:45AM +0000, Brian wrote:
> > #3410 downloaded this morning, unusually, very slowly. Any reason,
> > perhaps?
> 2016-02-16 09:36:12 (3.64 MB/s) - NetSurf-gcc--3410.zip' saved
> [3842257/3842257]
> Works for me.
> B.
#3413 (104kB/s) which works, sort of.
Re: Very slow download
On Tue, Feb 16, 2016 at 09:23:45AM +0000, Brian wrote:
> #3410 downloaded this morning, unusually, very slowly. Any reason, perhaps?
2016-02-16 09:36:12 (3.64 MB/s) - 'NetSurf-gcc--3410.zip' saved [3842257/3842257]
Works for me.
B.
> #3410 downloaded this morning, unusually, very slowly. Any reason, perhaps?
2016-02-16 09:36:12 (3.64 MB/s) - 'NetSurf-gcc--3410.zip' saved [3842257/3842257]
Works for me.
B.
Monday, 15 February 2016
[Rpcemu] RPCEmu Mac OS X test build of 0.8.14
I've made a test build of 0.8.14 for Mac, which you can find at:
https://bitbucket.org/caliston/rpcemu-spoon-caliston/downloads
I've also applied a patch from Sprow to support 4GB files.
I have only done the minimal amount of testing, but it seems to work for me.
Other than rebuilding, outstanding problems (eg lack of Retina support) will
remain.
Please let me know if there are any new bugs this has introduced.
Theo
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
https://bitbucket.org/caliston/rpcemu-spoon-caliston/downloads
I've also applied a patch from Sprow to support 4GB files.
I have only done the minimal amount of testing, but it seems to work for me.
Other than rebuilding, outstanding problems (eg lack of Retina support) will
remain.
Please let me know if there are any new bugs this has introduced.
Theo
_______________________________________________
Rpcemu mailing list
Rpcemu@riscos.info
http://www.riscos.info/cgi-bin/mailman/listinfo/rpcemu
Re: Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
On 12/02/16 17:46, Jim Nagel wrote:
> NetSurf (#3382) timed out when I tried to look at the Codethink site,
> but when I immediately tried again it displayed immediately.
We made a change to the fetch behavior some time ago that resulted
in us not waiting very long for a server to respond.
We've moved back to the old behavior in build CI #3406.
Cheers,
--
Michael Drake http://www.codethink.co.uk/
> NetSurf (#3382) timed out when I tried to look at the Codethink site,
> but when I immediately tried again it displayed immediately.
We made a change to the fetch behavior some time ago that resulted
in us not waiting very long for a server to respond.
We've moved back to the old behavior in build CI #3406.
Cheers,
--
Michael Drake http://www.codethink.co.uk/
Sunday, 14 February 2016
Re: Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
In article <d5b58a5155.harriet@blueyonder.co.uk>, Harriet Bazley
<lists@orange.wingsandbeaks.org.uk> wrote:
> On 12 Feb 2016 as I do recall, Michael Drake wrote:
> >
> > Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
> > ==============================================================
> >
> > Over the last couple of weeks I have been lucky enough to be able to
> > do work on the NetSurf project at work. At
> > [Codethink](http://www.codethink.co.uk/) we have an arrangement that
> > allows engineers to fill time between projects by making
> > contributions to certain open source projects.
> >
> > My request to work on [NetSurf](http://www.netsurf-browser.org/), and
> > more particularly, its [LibDOM](http://ns-b.org/projects/libdom/) and
> > [LibCSS](http://ns-b.org/projects/libcss/) sub-projects was approved
> > and I was able to get a good amount done.
> [snip]
> I'm impressed that so much of this seems to relate to making Netsurf
> faster and more efficient!
As someone who has been a little lax in keeping up with the bleeding edge
I'm really happy having downloaded #3403 to find it seems quicker than
the 'ancient' version I was using.
The spurious carriage returns which showed up inside tables seem to have
gone. Hurrah.
--
Tim Hill
www.timil.com
web sites * multimedia * training
<lists@orange.wingsandbeaks.org.uk> wrote:
> On 12 Feb 2016 as I do recall, Michael Drake wrote:
> >
> > Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
> > ==============================================================
> >
> > Over the last couple of weeks I have been lucky enough to be able to
> > do work on the NetSurf project at work. At
> > [Codethink](http://www.codethink.co.uk/) we have an arrangement that
> > allows engineers to fill time between projects by making
> > contributions to certain open source projects.
> >
> > My request to work on [NetSurf](http://www.netsurf-browser.org/), and
> > more particularly, its [LibDOM](http://ns-b.org/projects/libdom/) and
> > [LibCSS](http://ns-b.org/projects/libcss/) sub-projects was approved
> > and I was able to get a good amount done.
> [snip]
> I'm impressed that so much of this seems to relate to making Netsurf
> faster and more efficient!
As someone who has been a little lax in keeping up with the bleeding edge
I'm really happy having downloaded #3403 to find it seems quicker than
the 'ancient' version I was using.
The spurious carriage returns which showed up inside tables seem to have
gone. Hurrah.
--
Tim Hill
www.timil.com
web sites * multimedia * training
Saturday, 13 February 2016
Re: Backgrounds fill stopped working?
In article <56BFB079.6050400@netsurf-browser.org>,
Michael Drake <tlsa@netsurf-browser.org> wrote:
> On 13/02/16 16:08, Bob Latham wrote:
> > Is it me or has <body background="images/bgg.gif"> stopped working?
> Yep. Should be fixed in the next build, thanks.
Thank you.
Bob.
Michael Drake <tlsa@netsurf-browser.org> wrote:
> On 13/02/16 16:08, Bob Latham wrote:
> > Is it me or has <body background="images/bgg.gif"> stopped working?
> Yep. Should be fixed in the next build, thanks.
Thank you.
Bob.
Re: Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
On 12 Feb 2016 as I do recall,
Michael Drake wrote:
>
> Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
> ==============================================================
>
> Over the last couple of weeks I have been lucky enough to be able
> to do work on the NetSurf project at work.
> At [Codethink](http://www.codethink.co.uk/) we have an arrangement
> that allows engineers to fill time between projects by making
> contributions to certain open source projects.
>
> My request to work on [NetSurf](http://www.netsurf-browser.org/),
> and more particularly, its [LibDOM](http://ns-b.org/projects/libdom/)
> and [LibCSS](http://ns-b.org/projects/libcss/) sub-projects was
> approved and I was able to get a good amount done.
[snip]
I'm impressed that so much of this seems to relate to making Netsurf faster
and more efficient!
--
Harriet Bazley == Loyaulte me lie ==
It is easier to fight for one's principles than to live up to them.
Michael Drake wrote:
>
> Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
> ==============================================================
>
> Over the last couple of weeks I have been lucky enough to be able
> to do work on the NetSurf project at work.
> At [Codethink](http://www.codethink.co.uk/) we have an arrangement
> that allows engineers to fill time between projects by making
> contributions to certain open source projects.
>
> My request to work on [NetSurf](http://www.netsurf-browser.org/),
> and more particularly, its [LibDOM](http://ns-b.org/projects/libdom/)
> and [LibCSS](http://ns-b.org/projects/libcss/) sub-projects was
> approved and I was able to get a good amount done.
[snip]
I'm impressed that so much of this seems to relate to making Netsurf faster
and more efficient!
--
Harriet Bazley == Loyaulte me lie ==
It is easier to fight for one's principles than to live up to them.
Re: Backgrounds fill stopped working?
On 13/02/16 16:08, Bob Latham wrote:
> Is it me or has <body background="images/bgg.gif"> stopped working?
Yep. Should be fixed in the next build, thanks.
--
Michael Drake http://www.netsurf-browser.org/
> Is it me or has <body background="images/bgg.gif"> stopped working?
Yep. Should be fixed in the next build, thanks.
--
Michael Drake http://www.netsurf-browser.org/
Re: Backgrounds fill stopped working?
On Sat, 13 Feb 2016 16:08:57 +0000 (GMT), Bob Latham wrote:
> Is it me or has
>
> <body background="images/bgg.gif">
>
> stopped working? All of the pages on my Intranet according to Netsurf have
> gone white.
Just tried it here - I have an intranet page which specifies a
background image in CSS which is working. When I change it to specify
it in the body tag, it doesn't appear.
I will hazard a guess that this is related to the recent CSS hints
work. Please raise it on the bugtracker.
Chris
> Is it me or has
>
> <body background="images/bgg.gif">
>
> stopped working? All of the pages on my Intranet according to Netsurf have
> gone white.
Just tried it here - I have an intranet page which specifies a
background image in CSS which is working. When I change it to specify
it in the body tag, it doesn't appear.
I will hazard a guess that this is related to the recent CSS hints
work. Please raise it on the bugtracker.
Chris
Backgrounds fill stopped working?
Is it me or has
<body background="images/bgg.gif">
stopped working? All of the pages on my Intranet according to Netsurf have
gone white.
Probably me missing the obvious again. :-)
Bob.
<body background="images/bgg.gif">
stopped working? All of the pages on my Intranet according to Netsurf have
gone white.
Probably me missing the obvious again. :-)
Bob.
Friday, 12 February 2016
Re: Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
Excellent effort, Michael. Thanks from all of Us Users to you and
Codethink!
> ... At [Codethink](http://www.codethink.co.uk/) we have an arrangement
> that allows engineers to fill time between projects by making
> contributions to certain open-source projects.
Netsurf (#3382) timed out when I tried to look at the Codethink site,
but when I immediately tried again it displayed immediately.
Curious why that should be.
--
Jim Nagel www.archivemag.co.uk
Codethink!
> ... At [Codethink](http://www.codethink.co.uk/) we have an arrangement
> that allows engineers to fill time between projects by making
> contributions to certain open-source projects.
Netsurf (#3382) timed out when I tried to look at the Codethink site,
but when I immediately tried again it displayed immediately.
Curious why that should be.
--
Jim Nagel www.archivemag.co.uk
Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
Review: Codethink sponsored work on NetSurf, LibCSS and LibDOM
==============================================================
Over the last couple of weeks I have been lucky enough to be able
to do work on the NetSurf project at work.
At [Codethink](http://www.codethink.co.uk/) we have an arrangement
that allows engineers to fill time between projects by making
contributions to certain open source projects.
My request to work on [NetSurf](http://www.netsurf-browser.org/),
and more particularly, its [LibDOM](http://ns-b.org/projects/libdom/)
and [LibCSS](http://ns-b.org/projects/libcss/) sub-projects was
approved and I was able to get a good amount done.
LibDOM
------
Most of my time has been spent improving LibDOM; the Document Object
Model library. Since the recent moves towards adding JavaScript
support in NetSurf, various parts of LibDOM that had never been used
are beginning to get exercised. This has revealed various issues.
The first series of changes affected the specialisations of HTMLElement:
* Added a new, more efficient means of identifying an HTML element's
tag name.
* Fixed some missing HTML element specialisations.
* Made some optimisations relating to how the appropriate specialisation
for an element is determined.
* Made the internal workings of HTMLElement specialisation more consistent.
* Exposed a function to enabled client applications to use the new
efficient means of identifying HTML element tag types.
6d60021 Split out tag names from general string table and enum.
57 files changed, 502 insertions(+), 324 deletions(-)
3894e33 Split detection of HTML TAG type out into helper.
1 file changed, 325 insertions(+), 138 deletions(-)
1f07c35 Enable table row HTML element specialisation.
1 file changed, 4 insertions(+)
dd2e075 Add missing specialisation of BLOCKQUOTE.
1 file changed, 4 insertions(+)
8d09ec0 Add missing detection of HTMLDivElement specialisation.
1 file changed, 8 insertions(+)
81c012f Add missing detection of HTMLMetaElement specialisation.
1 file changed, 7 insertions(+)
e346fef Fix up HTMLBaseElement specialisation.
2 files changed, 8 insertions(+), 12 deletions(-)
13df5af Add missing detection of HTMLBaseElement specialisation.
1 file changed, 8 insertions(+)
0174d8b Remove default from HTML element specialisation switch.
1 file changed, 1 deletion(-)
12385d1 Optimise element specialisation: use non-caseless
comparison in ladder.
1 file changed, 63 insertions(+), 63 deletions(-)
13f0636 Optimise HTMLElement specialisation slightly.
1 file changed, 222 insertions(+), 189 deletions(-)
5b6c191 Fix longstanding failure to handle HTMLDirectoryElement
specialisation.
3 files changed, 6 insertions(+), 9 deletions(-)
9def0a4 Merge branch 'tlsa/html-element-type'
4bd00fd Fix: Pass tag_name through to HTMLQuoteElement initialiser.
3 files changed, 9 insertions(+), 8 deletions(-)
809a7dc Merge branch 'tlsa/html-element-type'
bab3f35 Simplified consistant interface to HTMLElement creation.
107 files changed, 894 insertions(+), 972 deletions(-)
5007c80 Store HTMLElement tag types as enum on the html elements.
6 files changed, 15 insertions(+), 7 deletions(-)
fa21110 Style: Fix indentation and tidy up.
1 file changed, 78 insertions(+), 66 deletions(-)
cedfbbc Add function to get html element's tag type.
4 files changed, 41 insertions(+), 1 deletion(-)
e137aa0 Merge branch 'tlsa/html-element-type'
Various profiling sessions indicated that the lower-casing of HTML
attributes was taking too long, so that was optimised.
a37713d Optimise dom_string_(toupper|tolower) functions.
1 file changed, 9 insertions(+), 31 deletions(-)
Another area we were spending too much time was in event dispatch.
Previously we would gather a list of nodes from the event target node
all the way back to the root node. Now we generate a list containing
nodes with registered listeners. In most cases, when there is no
scripting involved we can now avoid generating the list at all.
b4a245f Don't add target event to list of capture/bubbling event
targets.
1 file changed, 2 insertions(+), 4 deletions(-)
7e9681b Remove parameter documentation for non-existant parameter.
1 file changed, 1 deletion(-)
2196dd1 Optimise event dispatch.
1 file changed, 57 insertions(+), 20 deletions(-)
737cb99 Further optimise event dispatch.
1 file changed, 20 insertions(+), 5 deletions(-)
More changes to HTMLElement handling, mostly to extend the list of
known tags to include those that are not specialised in the DOM spec.
acedecb Sort HTML elements to make it easier to see what's there.
1 file changed, 44 insertions(+), 44 deletions(-)
c12b1ea Add full list of HTML5 elements to enum.
2 files changed, 363 insertions(+), 202 deletions(-)
d9c4dec Merge branch 'tlsa/html-element-type'
9ef22b6 Add CENTER to list of HTML elements.
2 files changed, 4 insertions(+)
e9257f8 Merge branch 'tlsa/html-element-type'
Various issues were identified with the HTMLElement class and
its specialisations' copy constructors. These were fixed in
the following series.
557678e Style: Fix sporadic use of spaces for indentation in
element copy constructor.
1 file changed, 5 insertions(+), 5 deletions(-)
b38b017 Split out element content copying from allocation in copy
constructor.
2 files changed, 41 insertions(+), 25 deletions(-)
89f73a0 Fix: Implement HTMLElement copy constructor.
2 files changed, 37 insertions(+), 1 deletion(-)
1ad8e7c Fix: Handle class list allocation failure in HTMLElement
copy constructor.
1 file changed, 13 insertions(+), 5 deletions(-)
e3dd4b2 Fix: Proper copy constructor for HTMLAnchorElement.
2 files changed, 41 insertions(+), 3 deletions(-)
49c0d92 Fix: Proper copy constructor for HTMLAppletElement.
2 files changed, 41 insertions(+), 3 deletions(-)
28f754a Fix: Proper copy constructor for HTMLAreaElement.
2 files changed, 41 insertions(+), 3 deletions(-)
d6017f1 Fix: Proper copy constructor for HTMLBaseElement.
2 files changed, 41 insertions(+), 3 deletions(-)
3542d8d Fix: Proper copy constructor for HTMLBasefontElement.
2 files changed, 41 insertions(+), 3 deletions(-)
1eb6cc5 Fix: Proper copy constructor for HTMLBodyElement.
2 files changed, 41 insertions(+), 3 deletions(-)
6c8d440 Fix: Proper copy constructor for HTMLBRElement.
2 files changed, 41 insertions(+), 3 deletions(-)
42bcd5d Provide generic copy constructor for HTMLButtonElement.
2 files changed, 41 insertions(+), 3 deletions(-)
bcc4b3e Fix: HTMLButtonElement's copy constructor copies its
specialised members.
1 file changed, 4 insertions(+)
43930a9 Fix: Proper copy constructor for HTMLDirectoryElement.
2 files changed, 41 insertions(+), 3 deletions(-)
9942e99 Fix: Proper copy constructor for HTMLDivElement.
2 files changed, 41 insertions(+), 3 deletions(-)
80873f9 Fix: Proper copy constructor for HTMLDListElement.
2 files changed, 41 insertions(+), 3 deletions(-)
b5c1e7c Fix: Proper copy constructor for HTMLFieldsetElement.
2 files changed, 41 insertions(+), 3 deletions(-)
663650a Fix: Proper copy constructor for HTMLFontElement.
2 files changed, 41 insertions(+), 3 deletions(-)
eef199e Fix: Proper copy constructor for HTMLFormElement.
2 files changed, 41 insertions(+), 3 deletions(-)
bd8ff18 Fix: Proper copy constructor for HTMLFrameElement.
2 files changed, 41 insertions(+), 3 deletions(-)
f4bec4a Fix: Proper copy constructor for HTMLFramesetElement.
2 files changed, 41 insertions(+), 3 deletions(-)
098842b Fix: Proper copy constructor for HTMLHeadElement.
2 files changed, 41 insertions(+), 3 deletions(-)
7eb3726 Fix: Proper copy constructor for HTMLHeadingElement.
2 files changed, 41 insertions(+), 3 deletions(-)
bdff6f9 Fix: Proper copy constructor for HTMLHRElement.
2 files changed, 41 insertions(+), 3 deletions(-)
840478f Fix: Proper copy constructor for HTMLHTMLElement.
2 files changed, 41 insertions(+), 3 deletions(-)
66a5830 Fix: Proper copy constructor for HTMLIframeElement.
2 files changed, 41 insertions(+), 3 deletions(-)
d3dd126 Fix: Proper copy constructor for HTMLImageElement.
2 files changed, 41 insertions(+), 3 deletions(-)
9f795a6 Provide generic copy constructor for HTMLInputElement.
2 files changed, 41 insertions(+), 3 deletions(-)
ec5f036 Fix: HTMLInputElement's copy constructor copies its
specialised members.
1 file changed, 12 insertions(+)
df4df17 Fix: Proper copy constructor for HTMLIsindexElement.
2 files changed, 41 insertions(+), 3 deletions(-)
07b7dd8 Fix: Proper copy constructor for HTMLLabelElement.
2 files changed, 41 insertions(+), 3 deletions(-)
87d9b3f Fix: Proper copy constructor for HTMLLegendElement.
2 files changed, 41 insertions(+), 3 deletions(-)
139250e Fix: Proper copy constructor for HTMLLiElement.
2 files changed, 41 insertions(+), 3 deletions(-)
9863b47 Fix: Proper copy constructor for HTMLLinkElement.
2 files changed, 41 insertions(+), 3 deletions(-)
c027180 Fix: Proper copy constructor for HTMLMapElement.
2 files changed, 41 insertions(+), 3 deletions(-)
13c7b9e Cleanup: Remove odd declaration of callback outside header
guard.
2 files changed, 1 insertion(+), 3 deletions(-)
0238cec Fix: Proper copy constructor for HTMLMenuElement.
2 files changed, 41 insertions(+), 3 deletions(-)
a730440 Fix: Proper copy constructor for HTMLMetaElement.
2 files changed, 41 insertions(+), 3 deletions(-)
5ab7524 Fix: Proper copy constructor for HTMLModElement.
2 files changed, 41 insertions(+), 3 deletions(-)
f97e745 Fix: Proper copy constructor for HTMLObjectElement.
2 files changed, 41 insertions(+), 3 deletions(-)
84f9b4e Fix: Proper copy constructor for HTMLOListElement.
2 files changed, 41 insertions(+), 3 deletions(-)
64833e7 Fix: Proper copy constructor for HTMLOptGroupElement.
2 files changed, 41 insertions(+), 3 deletions(-)
89171f6 Provide generic copy constructor for HTMLOptionElement.
2 files changed, 41 insertions(+), 3 deletions(-)
37eb480 Fix: HTMLOptionElement's copy constructor copies its
specialised members.
1 file changed, 3 insertions(+)
842deef Fix: Proper copy constructor for HTMLParagraphElement.
2 files changed, 41 insertions(+), 3 deletions(-)
37692b4 Fix: Proper copy constructor for HTMLParamElement.
2 files changed, 41 insertions(+), 3 deletions(-)
c4b0e2a Fix: Proper copy constructor for HTMLPreElement.
2 files changed, 41 insertions(+), 3 deletions(-)
c4159ca Fix: Proper copy constructor for HTMLQuoteElement.
2 files changed, 41 insertions(+), 3 deletions(-)
d82f237 Fix: Proper copy constructor for HTMLScriptElement.
2 files changed, 41 insertions(+), 3 deletions(-)
cff37d7 Provide generic copy constructor for HTMLSelectElement.
2 files changed, 41 insertions(+), 3 deletions(-)
3c056b4 Fix: HTMLSelectElement's copy constructor copies its
specialised members.
1 file changed, 6 insertions(+)
2dda1dd Fix: Proper copy constructor for HTMLStyleElement.
2 files changed, 41 insertions(+), 3 deletions(-)
609e19e Fix: Proper copy constructor for HTMLTableElement.
2 files changed, 41 insertions(+), 3 deletions(-)
7782715 Cleanup: Remove odd declaration of callbacks outside header
guard.
2 files changed, 2 insertions(+), 4 deletions(-)
103c582 Cleanup: Don't put static function in header.
2 files changed, 2 insertions(+), 5 deletions(-)
349992c Fix: Proper copy constructor for HTMLTableCaptionElement.
2 files changed, 41 insertions(+), 3 deletions(-)
7d0c256 Fix: Proper copy constructor for HTMLTableCellElement.
2 files changed, 41 insertions(+), 3 deletions(-)
41fd1fe Fix: Proper copy constructor for HTMLTableColElement.
2 files changed, 41 insertions(+), 3 deletions(-)
e13599e Fix: Proper copy constructor for HTMLTableRowElement.
2 files changed, 41 insertions(+), 3 deletions(-)
f566b7c Cleanup: Remove odd declaration of callback outside header
guard.
2 files changed, 1 insertion(+), 2 deletions(-)
afec576 Fix: Proper copy constructor for HTMLTableSectionElement.
2 files changed, 41 insertions(+), 3 deletions(-)
848c41d Cleanup: Remove odd declaration of callback outside header
guard.
2 files changed, 1 insertion(+), 2 deletions(-)
b26b3cf Provide generic copy constructor for HTMLTextAreaElement.
2 files changed, 41 insertions(+), 3 deletions(-)
0480b0a Fix: HTMLTextAreaElement's copy constructor copies its
specialised members.
1 file changed, 9 insertions(+)
0f69427 Fix: Proper copy constructor for HTMLTitleElement.
2 files changed, 41 insertions(+), 3 deletions(-)
d19eda0 Fix: Proper copy constructor for HTMLUListElement.
2 files changed, 41 insertions(+), 3 deletions(-)
68c4463 Merge branch 'tlsa/fix-html-element-copy-constructors'
Finally a Coverity static analysis issue was fixed.
9f2c89e Fix Coverity #1350096.
1 file changed, 1 insertion(+), 1 deletion(-)
Overall, the there were 99 commits to LibDOM over the course of this stint.
LibCSS
------
Most of the work done on LibCSS was cherry-picking a commit to change
presentational hint handling to be more efficient from another of my
branches, and getting it merged. This was dependent on getting NetSurf
ready to handle the change.
This change to presentational hint handling has made the selection
process much faster for old-school sites that use HTML attributes
for styling content, rather than CSS. It also speeds up the handling
of modern CSS-using sites too, as there is now no overhead when there
are no hints present in the document.
d4b2c5c Merge branch 'tlsa/upfront-hints'
9024333 Documentation: Presentational hint API change.
1 file changed, 19 insertions(+)
df3d8e8 Documentation: Fix typos.
1 file changed, 3 insertions(+), 3 deletions(-)
NetSurf
-------
Much of the work done on NetSurf was in response to the changes made
in LibCSS and LibDOM, but I also spent time on bug fixes, cleanups,
and pre-release testing.
The first set of changes were mostly related to JavaScript creation
of HTMLElements, and making use of the new LibDOM API for determining
an element's tag type.
b58176c Spaces to tabs.
1 file changed, 1 insertion(+), 1 deletion(-)
b002ba7 Simplify special element handling in node inserted callback.
1 file changed, 31 insertions(+), 31 deletions(-)
6904622 JavaScript: Fix & optimise HTMLElement specialisation proto
str generation.
1 file changed, 199 insertions(+), 9 deletions(-)
2443cc4 Attempt to squash warning that happens on CI.
1 file changed, 3 insertions(+), 3 deletions(-)
b1dbc04 Add paragraph proto test.
1 file changed, 23 insertions(+)
9fb755b Handle known HTML elements without specialisations.
1 file changed, 7 insertions(+), 1 deletion(-)
3cc80b6 Add test for document.write() adding a STYLE element.
1 file changed, 15 insertions(+)
Next was a complete rewrite of CSS presentational hints handling.
Presentational hints are mostly legacy HTML element attributes which
affect the presentation of the elements. They affect the CSS computed
style when no other styles have affected it. Using the new LibCSS API
coupled with the new LibDOM element tag type feature has made support
for presentational hints much faster.
2784514 Rewrite hints handling to be more efficient.
2 files changed, 788 insertions(+), 1123 deletions(-)
eb67607 Merge branch 'tlsa/upfront-hints'
I noticed and investigated an apparent memory leak. Some changes were
made to improve the cache logging, which ruled the cache out as the
cause of the problem. The leak was tracked to a particular issue in
the GTK front end's text rendering.
63fb2cb Add URL to llcache object destruction logging.
1 file changed, 2 insertions(+), 1 deletion(-)
fc2d766 Style: Wrap some long lines.
1 file changed, 31 insertions(+), 18 deletions(-)
255361a Log the cache limit along side size, after clean.
1 file changed, 1 insertion(+), 1 deletion(-)
9cb522c Don't haemorrhage Pango layouts out of nsfont_paint.
1 file changed, 1 insertion(+)
A few tidy-ups to GTK font rendering and core CSS hint handling:
df2d62a Use readonly get_line, since it is supposed to be faster.
1 file changed, 1 insertion(+), 1 deletion(-)
f5be1eb Add switch case fallthrough comments.
1 file changed, 5 insertions(+)
5108424 Add switch case fallthrough comment.
1 file changed, 1 insertion(+)
3af77ea Can free pango font desc as soon as it's set.
1 file changed, 12 insertions(+), 10 deletions(-)
Finally I fixed several bugs on the bug tracker.
The first was an assertion (crash) that could occur during HTML
layout. The issue stemmed from NetSurf's allowing :before and
:after pseudo elements to be generated for replaced elements.
This behaviour was unspecified in CSS 2.1, but in CSS 3 it is
not permitted.
Second was to remove some code which was complicating the status
bar updating for HTML contents.
Third was to fix the object accounting that is done to keep track
of fetches used for additional resources such as images when
rendering an HTML content. This accounting went wrong when a
fetch for an HTML content was stopped, preventing the content
ever reaching the DONE state, and causing misleading status
bar messages.
3bee7b7 Layout: Don't generate :before and :after boxes for
replaced elements.
2 files changed, 17 insertions(+), 6 deletions(-)
d4a01d5 HTML: Remove some status bar updating code.
4 files changed, 43 deletions(-)
04e61b7 Fix object accounting for aborted HTML contents.
1 file changed, 4 insertions(+), 3 deletions(-)
I also did quite a bit of testing, particularly of the JavaScript
related code, to ensure that there is nothing major blocking the
forthcoming NetSurf 3.4 release.
Overall, there were 20 commits to NetSurf through the course of this stint.
Overview and thanks
-------------------
I was able to get many hours of work done on the NetSurf project over
this period. It has been a very nice experience, working on this
project at work. Thanks to the work environment, I found myself more
focused than I usually am when I'm working on it in my spare time.
Many thanks to [Codethink](http://www.codethink.co.uk/) for providing
this opportunity. I'm very grateful and I've really enjoyed it.
The stats and logs presented were gathered using the following commands
in the NetSurf, LibDOM and LibCSS repositories.
$ git shortlog -n -s -e | grep "michael.drake@codethink.co.uk"
$ git log --author="Michael Drake <michael.drake@codethink.co.uk>"
--oneline --shortstat --reverse
--
Michael Drake http://www.codethink.co.uk/
==============================================================
Over the last couple of weeks I have been lucky enough to be able
to do work on the NetSurf project at work.
At [Codethink](http://www.codethink.co.uk/) we have an arrangement
that allows engineers to fill time between projects by making
contributions to certain open source projects.
My request to work on [NetSurf](http://www.netsurf-browser.org/),
and more particularly, its [LibDOM](http://ns-b.org/projects/libdom/)
and [LibCSS](http://ns-b.org/projects/libcss/) sub-projects was
approved and I was able to get a good amount done.
LibDOM
------
Most of my time has been spent improving LibDOM; the Document Object
Model library. Since the recent moves towards adding JavaScript
support in NetSurf, various parts of LibDOM that had never been used
are beginning to get exercised. This has revealed various issues.
The first series of changes affected the specialisations of HTMLElement:
* Added a new, more efficient means of identifying an HTML element's
tag name.
* Fixed some missing HTML element specialisations.
* Made some optimisations relating to how the appropriate specialisation
for an element is determined.
* Made the internal workings of HTMLElement specialisation more consistent.
* Exposed a function to enabled client applications to use the new
efficient means of identifying HTML element tag types.
6d60021 Split out tag names from general string table and enum.
57 files changed, 502 insertions(+), 324 deletions(-)
3894e33 Split detection of HTML TAG type out into helper.
1 file changed, 325 insertions(+), 138 deletions(-)
1f07c35 Enable table row HTML element specialisation.
1 file changed, 4 insertions(+)
dd2e075 Add missing specialisation of BLOCKQUOTE.
1 file changed, 4 insertions(+)
8d09ec0 Add missing detection of HTMLDivElement specialisation.
1 file changed, 8 insertions(+)
81c012f Add missing detection of HTMLMetaElement specialisation.
1 file changed, 7 insertions(+)
e346fef Fix up HTMLBaseElement specialisation.
2 files changed, 8 insertions(+), 12 deletions(-)
13df5af Add missing detection of HTMLBaseElement specialisation.
1 file changed, 8 insertions(+)
0174d8b Remove default from HTML element specialisation switch.
1 file changed, 1 deletion(-)
12385d1 Optimise element specialisation: use non-caseless
comparison in ladder.
1 file changed, 63 insertions(+), 63 deletions(-)
13f0636 Optimise HTMLElement specialisation slightly.
1 file changed, 222 insertions(+), 189 deletions(-)
5b6c191 Fix longstanding failure to handle HTMLDirectoryElement
specialisation.
3 files changed, 6 insertions(+), 9 deletions(-)
9def0a4 Merge branch 'tlsa/html-element-type'
4bd00fd Fix: Pass tag_name through to HTMLQuoteElement initialiser.
3 files changed, 9 insertions(+), 8 deletions(-)
809a7dc Merge branch 'tlsa/html-element-type'
bab3f35 Simplified consistant interface to HTMLElement creation.
107 files changed, 894 insertions(+), 972 deletions(-)
5007c80 Store HTMLElement tag types as enum on the html elements.
6 files changed, 15 insertions(+), 7 deletions(-)
fa21110 Style: Fix indentation and tidy up.
1 file changed, 78 insertions(+), 66 deletions(-)
cedfbbc Add function to get html element's tag type.
4 files changed, 41 insertions(+), 1 deletion(-)
e137aa0 Merge branch 'tlsa/html-element-type'
Various profiling sessions indicated that the lower-casing of HTML
attributes was taking too long, so that was optimised.
a37713d Optimise dom_string_(toupper|tolower) functions.
1 file changed, 9 insertions(+), 31 deletions(-)
Another area we were spending too much time was in event dispatch.
Previously we would gather a list of nodes from the event target node
all the way back to the root node. Now we generate a list containing
nodes with registered listeners. In most cases, when there is no
scripting involved we can now avoid generating the list at all.
b4a245f Don't add target event to list of capture/bubbling event
targets.
1 file changed, 2 insertions(+), 4 deletions(-)
7e9681b Remove parameter documentation for non-existant parameter.
1 file changed, 1 deletion(-)
2196dd1 Optimise event dispatch.
1 file changed, 57 insertions(+), 20 deletions(-)
737cb99 Further optimise event dispatch.
1 file changed, 20 insertions(+), 5 deletions(-)
More changes to HTMLElement handling, mostly to extend the list of
known tags to include those that are not specialised in the DOM spec.
acedecb Sort HTML elements to make it easier to see what's there.
1 file changed, 44 insertions(+), 44 deletions(-)
c12b1ea Add full list of HTML5 elements to enum.
2 files changed, 363 insertions(+), 202 deletions(-)
d9c4dec Merge branch 'tlsa/html-element-type'
9ef22b6 Add CENTER to list of HTML elements.
2 files changed, 4 insertions(+)
e9257f8 Merge branch 'tlsa/html-element-type'
Various issues were identified with the HTMLElement class and
its specialisations' copy constructors. These were fixed in
the following series.
557678e Style: Fix sporadic use of spaces for indentation in
element copy constructor.
1 file changed, 5 insertions(+), 5 deletions(-)
b38b017 Split out element content copying from allocation in copy
constructor.
2 files changed, 41 insertions(+), 25 deletions(-)
89f73a0 Fix: Implement HTMLElement copy constructor.
2 files changed, 37 insertions(+), 1 deletion(-)
1ad8e7c Fix: Handle class list allocation failure in HTMLElement
copy constructor.
1 file changed, 13 insertions(+), 5 deletions(-)
e3dd4b2 Fix: Proper copy constructor for HTMLAnchorElement.
2 files changed, 41 insertions(+), 3 deletions(-)
49c0d92 Fix: Proper copy constructor for HTMLAppletElement.
2 files changed, 41 insertions(+), 3 deletions(-)
28f754a Fix: Proper copy constructor for HTMLAreaElement.
2 files changed, 41 insertions(+), 3 deletions(-)
d6017f1 Fix: Proper copy constructor for HTMLBaseElement.
2 files changed, 41 insertions(+), 3 deletions(-)
3542d8d Fix: Proper copy constructor for HTMLBasefontElement.
2 files changed, 41 insertions(+), 3 deletions(-)
1eb6cc5 Fix: Proper copy constructor for HTMLBodyElement.
2 files changed, 41 insertions(+), 3 deletions(-)
6c8d440 Fix: Proper copy constructor for HTMLBRElement.
2 files changed, 41 insertions(+), 3 deletions(-)
42bcd5d Provide generic copy constructor for HTMLButtonElement.
2 files changed, 41 insertions(+), 3 deletions(-)
bcc4b3e Fix: HTMLButtonElement's copy constructor copies its
specialised members.
1 file changed, 4 insertions(+)
43930a9 Fix: Proper copy constructor for HTMLDirectoryElement.
2 files changed, 41 insertions(+), 3 deletions(-)
9942e99 Fix: Proper copy constructor for HTMLDivElement.
2 files changed, 41 insertions(+), 3 deletions(-)
80873f9 Fix: Proper copy constructor for HTMLDListElement.
2 files changed, 41 insertions(+), 3 deletions(-)
b5c1e7c Fix: Proper copy constructor for HTMLFieldsetElement.
2 files changed, 41 insertions(+), 3 deletions(-)
663650a Fix: Proper copy constructor for HTMLFontElement.
2 files changed, 41 insertions(+), 3 deletions(-)
eef199e Fix: Proper copy constructor for HTMLFormElement.
2 files changed, 41 insertions(+), 3 deletions(-)
bd8ff18 Fix: Proper copy constructor for HTMLFrameElement.
2 files changed, 41 insertions(+), 3 deletions(-)
f4bec4a Fix: Proper copy constructor for HTMLFramesetElement.
2 files changed, 41 insertions(+), 3 deletions(-)
098842b Fix: Proper copy constructor for HTMLHeadElement.
2 files changed, 41 insertions(+), 3 deletions(-)
7eb3726 Fix: Proper copy constructor for HTMLHeadingElement.
2 files changed, 41 insertions(+), 3 deletions(-)
bdff6f9 Fix: Proper copy constructor for HTMLHRElement.
2 files changed, 41 insertions(+), 3 deletions(-)
840478f Fix: Proper copy constructor for HTMLHTMLElement.
2 files changed, 41 insertions(+), 3 deletions(-)
66a5830 Fix: Proper copy constructor for HTMLIframeElement.
2 files changed, 41 insertions(+), 3 deletions(-)
d3dd126 Fix: Proper copy constructor for HTMLImageElement.
2 files changed, 41 insertions(+), 3 deletions(-)
9f795a6 Provide generic copy constructor for HTMLInputElement.
2 files changed, 41 insertions(+), 3 deletions(-)
ec5f036 Fix: HTMLInputElement's copy constructor copies its
specialised members.
1 file changed, 12 insertions(+)
df4df17 Fix: Proper copy constructor for HTMLIsindexElement.
2 files changed, 41 insertions(+), 3 deletions(-)
07b7dd8 Fix: Proper copy constructor for HTMLLabelElement.
2 files changed, 41 insertions(+), 3 deletions(-)
87d9b3f Fix: Proper copy constructor for HTMLLegendElement.
2 files changed, 41 insertions(+), 3 deletions(-)
139250e Fix: Proper copy constructor for HTMLLiElement.
2 files changed, 41 insertions(+), 3 deletions(-)
9863b47 Fix: Proper copy constructor for HTMLLinkElement.
2 files changed, 41 insertions(+), 3 deletions(-)
c027180 Fix: Proper copy constructor for HTMLMapElement.
2 files changed, 41 insertions(+), 3 deletions(-)
13c7b9e Cleanup: Remove odd declaration of callback outside header
guard.
2 files changed, 1 insertion(+), 3 deletions(-)
0238cec Fix: Proper copy constructor for HTMLMenuElement.
2 files changed, 41 insertions(+), 3 deletions(-)
a730440 Fix: Proper copy constructor for HTMLMetaElement.
2 files changed, 41 insertions(+), 3 deletions(-)
5ab7524 Fix: Proper copy constructor for HTMLModElement.
2 files changed, 41 insertions(+), 3 deletions(-)
f97e745 Fix: Proper copy constructor for HTMLObjectElement.
2 files changed, 41 insertions(+), 3 deletions(-)
84f9b4e Fix: Proper copy constructor for HTMLOListElement.
2 files changed, 41 insertions(+), 3 deletions(-)
64833e7 Fix: Proper copy constructor for HTMLOptGroupElement.
2 files changed, 41 insertions(+), 3 deletions(-)
89171f6 Provide generic copy constructor for HTMLOptionElement.
2 files changed, 41 insertions(+), 3 deletions(-)
37eb480 Fix: HTMLOptionElement's copy constructor copies its
specialised members.
1 file changed, 3 insertions(+)
842deef Fix: Proper copy constructor for HTMLParagraphElement.
2 files changed, 41 insertions(+), 3 deletions(-)
37692b4 Fix: Proper copy constructor for HTMLParamElement.
2 files changed, 41 insertions(+), 3 deletions(-)
c4b0e2a Fix: Proper copy constructor for HTMLPreElement.
2 files changed, 41 insertions(+), 3 deletions(-)
c4159ca Fix: Proper copy constructor for HTMLQuoteElement.
2 files changed, 41 insertions(+), 3 deletions(-)
d82f237 Fix: Proper copy constructor for HTMLScriptElement.
2 files changed, 41 insertions(+), 3 deletions(-)
cff37d7 Provide generic copy constructor for HTMLSelectElement.
2 files changed, 41 insertions(+), 3 deletions(-)
3c056b4 Fix: HTMLSelectElement's copy constructor copies its
specialised members.
1 file changed, 6 insertions(+)
2dda1dd Fix: Proper copy constructor for HTMLStyleElement.
2 files changed, 41 insertions(+), 3 deletions(-)
609e19e Fix: Proper copy constructor for HTMLTableElement.
2 files changed, 41 insertions(+), 3 deletions(-)
7782715 Cleanup: Remove odd declaration of callbacks outside header
guard.
2 files changed, 2 insertions(+), 4 deletions(-)
103c582 Cleanup: Don't put static function in header.
2 files changed, 2 insertions(+), 5 deletions(-)
349992c Fix: Proper copy constructor for HTMLTableCaptionElement.
2 files changed, 41 insertions(+), 3 deletions(-)
7d0c256 Fix: Proper copy constructor for HTMLTableCellElement.
2 files changed, 41 insertions(+), 3 deletions(-)
41fd1fe Fix: Proper copy constructor for HTMLTableColElement.
2 files changed, 41 insertions(+), 3 deletions(-)
e13599e Fix: Proper copy constructor for HTMLTableRowElement.
2 files changed, 41 insertions(+), 3 deletions(-)
f566b7c Cleanup: Remove odd declaration of callback outside header
guard.
2 files changed, 1 insertion(+), 2 deletions(-)
afec576 Fix: Proper copy constructor for HTMLTableSectionElement.
2 files changed, 41 insertions(+), 3 deletions(-)
848c41d Cleanup: Remove odd declaration of callback outside header
guard.
2 files changed, 1 insertion(+), 2 deletions(-)
b26b3cf Provide generic copy constructor for HTMLTextAreaElement.
2 files changed, 41 insertions(+), 3 deletions(-)
0480b0a Fix: HTMLTextAreaElement's copy constructor copies its
specialised members.
1 file changed, 9 insertions(+)
0f69427 Fix: Proper copy constructor for HTMLTitleElement.
2 files changed, 41 insertions(+), 3 deletions(-)
d19eda0 Fix: Proper copy constructor for HTMLUListElement.
2 files changed, 41 insertions(+), 3 deletions(-)
68c4463 Merge branch 'tlsa/fix-html-element-copy-constructors'
Finally a Coverity static analysis issue was fixed.
9f2c89e Fix Coverity #1350096.
1 file changed, 1 insertion(+), 1 deletion(-)
Overall, the there were 99 commits to LibDOM over the course of this stint.
LibCSS
------
Most of the work done on LibCSS was cherry-picking a commit to change
presentational hint handling to be more efficient from another of my
branches, and getting it merged. This was dependent on getting NetSurf
ready to handle the change.
This change to presentational hint handling has made the selection
process much faster for old-school sites that use HTML attributes
for styling content, rather than CSS. It also speeds up the handling
of modern CSS-using sites too, as there is now no overhead when there
are no hints present in the document.
d4b2c5c Merge branch 'tlsa/upfront-hints'
9024333 Documentation: Presentational hint API change.
1 file changed, 19 insertions(+)
df3d8e8 Documentation: Fix typos.
1 file changed, 3 insertions(+), 3 deletions(-)
NetSurf
-------
Much of the work done on NetSurf was in response to the changes made
in LibCSS and LibDOM, but I also spent time on bug fixes, cleanups,
and pre-release testing.
The first set of changes were mostly related to JavaScript creation
of HTMLElements, and making use of the new LibDOM API for determining
an element's tag type.
b58176c Spaces to tabs.
1 file changed, 1 insertion(+), 1 deletion(-)
b002ba7 Simplify special element handling in node inserted callback.
1 file changed, 31 insertions(+), 31 deletions(-)
6904622 JavaScript: Fix & optimise HTMLElement specialisation proto
str generation.
1 file changed, 199 insertions(+), 9 deletions(-)
2443cc4 Attempt to squash warning that happens on CI.
1 file changed, 3 insertions(+), 3 deletions(-)
b1dbc04 Add paragraph proto test.
1 file changed, 23 insertions(+)
9fb755b Handle known HTML elements without specialisations.
1 file changed, 7 insertions(+), 1 deletion(-)
3cc80b6 Add test for document.write() adding a STYLE element.
1 file changed, 15 insertions(+)
Next was a complete rewrite of CSS presentational hints handling.
Presentational hints are mostly legacy HTML element attributes which
affect the presentation of the elements. They affect the CSS computed
style when no other styles have affected it. Using the new LibCSS API
coupled with the new LibDOM element tag type feature has made support
for presentational hints much faster.
2784514 Rewrite hints handling to be more efficient.
2 files changed, 788 insertions(+), 1123 deletions(-)
eb67607 Merge branch 'tlsa/upfront-hints'
I noticed and investigated an apparent memory leak. Some changes were
made to improve the cache logging, which ruled the cache out as the
cause of the problem. The leak was tracked to a particular issue in
the GTK front end's text rendering.
63fb2cb Add URL to llcache object destruction logging.
1 file changed, 2 insertions(+), 1 deletion(-)
fc2d766 Style: Wrap some long lines.
1 file changed, 31 insertions(+), 18 deletions(-)
255361a Log the cache limit along side size, after clean.
1 file changed, 1 insertion(+), 1 deletion(-)
9cb522c Don't haemorrhage Pango layouts out of nsfont_paint.
1 file changed, 1 insertion(+)
A few tidy-ups to GTK font rendering and core CSS hint handling:
df2d62a Use readonly get_line, since it is supposed to be faster.
1 file changed, 1 insertion(+), 1 deletion(-)
f5be1eb Add switch case fallthrough comments.
1 file changed, 5 insertions(+)
5108424 Add switch case fallthrough comment.
1 file changed, 1 insertion(+)
3af77ea Can free pango font desc as soon as it's set.
1 file changed, 12 insertions(+), 10 deletions(-)
Finally I fixed several bugs on the bug tracker.
The first was an assertion (crash) that could occur during HTML
layout. The issue stemmed from NetSurf's allowing :before and
:after pseudo elements to be generated for replaced elements.
This behaviour was unspecified in CSS 2.1, but in CSS 3 it is
not permitted.
Second was to remove some code which was complicating the status
bar updating for HTML contents.
Third was to fix the object accounting that is done to keep track
of fetches used for additional resources such as images when
rendering an HTML content. This accounting went wrong when a
fetch for an HTML content was stopped, preventing the content
ever reaching the DONE state, and causing misleading status
bar messages.
3bee7b7 Layout: Don't generate :before and :after boxes for
replaced elements.
2 files changed, 17 insertions(+), 6 deletions(-)
d4a01d5 HTML: Remove some status bar updating code.
4 files changed, 43 deletions(-)
04e61b7 Fix object accounting for aborted HTML contents.
1 file changed, 4 insertions(+), 3 deletions(-)
I also did quite a bit of testing, particularly of the JavaScript
related code, to ensure that there is nothing major blocking the
forthcoming NetSurf 3.4 release.
Overall, there were 20 commits to NetSurf through the course of this stint.
Overview and thanks
-------------------
I was able to get many hours of work done on the NetSurf project over
this period. It has been a very nice experience, working on this
project at work. Thanks to the work environment, I found myself more
focused than I usually am when I'm working on it in my spare time.
Many thanks to [Codethink](http://www.codethink.co.uk/) for providing
this opportunity. I'm very grateful and I've really enjoyed it.
The stats and logs presented were gathered using the following commands
in the NetSurf, LibDOM and LibCSS repositories.
$ git shortlog -n -s -e | grep "michael.drake@codethink.co.uk"
$ git log --author="Michael Drake <michael.drake@codethink.co.uk>"
--oneline --shortstat --reverse
--
Michael Drake http://www.codethink.co.uk/
Subscribe to:
Posts (Atom)