You can subscribe to this list here.
2001 |
Jan
|
Feb
(20) |
Mar
(29) |
Apr
(10) |
May
(10) |
Jun
(7) |
Jul
(6) |
Aug
(59) |
Sep
(19) |
Oct
(55) |
Nov
(22) |
Dec
(40) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2002 |
Jan
(56) |
Feb
(71) |
Mar
(179) |
Apr
(41) |
May
(26) |
Jun
(52) |
Jul
(62) |
Aug
(19) |
Sep
(87) |
Oct
(188) |
Nov
(95) |
Dec
(30) |
2003 |
Jan
(83) |
Feb
(119) |
Mar
(174) |
Apr
(77) |
May
(85) |
Jun
(52) |
Jul
(67) |
Aug
(121) |
Sep
(147) |
Oct
(96) |
Nov
(89) |
Dec
(144) |
2004 |
Jan
(92) |
Feb
(172) |
Mar
(205) |
Apr
(201) |
May
(105) |
Jun
(42) |
Jul
(94) |
Aug
(109) |
Sep
(81) |
Oct
(59) |
Nov
(84) |
Dec
(68) |
2005 |
Jan
(56) |
Feb
(57) |
Mar
(183) |
Apr
(139) |
May
(131) |
Jun
(178) |
Jul
(62) |
Aug
(42) |
Sep
(95) |
Oct
(47) |
Nov
(73) |
Dec
(47) |
2006 |
Jan
(66) |
Feb
(31) |
Mar
(51) |
Apr
(20) |
May
(49) |
Jun
(26) |
Jul
(23) |
Aug
(65) |
Sep
(67) |
Oct
(26) |
Nov
(16) |
Dec
(8) |
2007 |
Jan
(18) |
Feb
(43) |
Mar
(43) |
Apr
(16) |
May
(33) |
Jun
(48) |
Jul
(34) |
Aug
(7) |
Sep
(9) |
Oct
(55) |
Nov
(44) |
Dec
(73) |
2008 |
Jan
(37) |
Feb
(97) |
Mar
(44) |
Apr
(33) |
May
(79) |
Jun
(11) |
Jul
(66) |
Aug
(9) |
Sep
(12) |
Oct
(6) |
Nov
(12) |
Dec
(19) |
2009 |
Jan
(12) |
Feb
(13) |
Mar
(19) |
Apr
(30) |
May
(59) |
Jun
(22) |
Jul
(11) |
Aug
(59) |
Sep
(82) |
Oct
(25) |
Nov
(51) |
Dec
(27) |
2010 |
Jan
(27) |
Feb
(8) |
Mar
(29) |
Apr
(9) |
May
(39) |
Jun
(6) |
Jul
(8) |
Aug
(22) |
Sep
(33) |
Oct
(8) |
Nov
(35) |
Dec
(9) |
2011 |
Jan
(62) |
Feb
(19) |
Mar
(31) |
Apr
(19) |
May
(1) |
Jun
(1) |
Jul
(17) |
Aug
(10) |
Sep
(14) |
Oct
(11) |
Nov
|
Dec
|
2012 |
Jan
(1) |
Feb
(11) |
Mar
|
Apr
(1) |
May
(5) |
Jun
(7) |
Jul
(22) |
Aug
(22) |
Sep
(30) |
Oct
(23) |
Nov
(19) |
Dec
|
2013 |
Jan
(6) |
Feb
(1) |
Mar
(10) |
Apr
(7) |
May
(3) |
Jun
(3) |
Jul
|
Aug
(3) |
Sep
(9) |
Oct
(14) |
Nov
(9) |
Dec
(5) |
2014 |
Jan
(13) |
Feb
(1) |
Mar
(6) |
Apr
(3) |
May
(5) |
Jun
(2) |
Jul
(20) |
Aug
(6) |
Sep
(26) |
Oct
(25) |
Nov
(20) |
Dec
(41) |
2015 |
Jan
(9) |
Feb
(35) |
Mar
(9) |
Apr
(28) |
May
(20) |
Jun
(3) |
Jul
(5) |
Aug
|
Sep
(2) |
Oct
(4) |
Nov
|
Dec
(3) |
2016 |
Jan
|
Feb
|
Mar
|
Apr
(5) |
May
(12) |
Jun
(35) |
Jul
(1) |
Aug
|
Sep
(1) |
Oct
|
Nov
|
Dec
(7) |
2017 |
Jan
(28) |
Feb
(14) |
Mar
(4) |
Apr
(5) |
May
(4) |
Jun
(2) |
Jul
|
Aug
(1) |
Sep
|
Oct
(3) |
Nov
|
Dec
(8) |
2018 |
Jan
|
Feb
(1) |
Mar
(3) |
Apr
(1) |
May
(1) |
Jun
(3) |
Jul
(3) |
Aug
|
Sep
|
Oct
(1) |
Nov
(1) |
Dec
|
2019 |
Jan
|
Feb
|
Mar
|
Apr
(3) |
May
(7) |
Jun
(2) |
Jul
|
Aug
(1) |
Sep
(2) |
Oct
(3) |
Nov
(7) |
Dec
|
2020 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(4) |
Jun
|
Jul
(10) |
Aug
(3) |
Sep
|
Oct
(1) |
Nov
|
Dec
|
2021 |
Jan
|
Feb
|
Mar
(4) |
Apr
(21) |
May
(8) |
Jun
(3) |
Jul
|
Aug
|
Sep
(1) |
Oct
(10) |
Nov
|
Dec
|
2022 |
Jan
(1) |
Feb
(4) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(4) |
Sep
|
Oct
|
Nov
|
Dec
|
2023 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(5) |
Sep
|
Oct
|
Nov
|
Dec
|
2024 |
Jan
(3) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(3) |
Aug
|
Sep
(7) |
Oct
|
Nov
|
Dec
|
2025 |
Jan
(3) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Tim J. <tj...@to...> - 2011-02-09 23:33:52
|
Hi Kevin, Your native toolbars seem to be missing... Tim On Feb 9, 2011, at 4:14 PM, Kevin Walzer wrote: > By way of comparison, here is the native toolbar: > > http://www.codebykevin.com/cocoa-toolbar.png > > > and the emulated one: > > http://www.codebykevin.com/tk-unified-latest.png > > --K > > On 2/9/11 5:59 PM, Kevin Walzer wrote: >> I've been doing a bit of hacking with Tk again, and I've discovered how >> to enable a real NSToolbar in a Tk window via a Tk extension. Here's a >> screenshot: >> >> http://www.codebykevin.com/cocoa-toolbar.png >> >> This is in the very early stages, but I think it will be feasible for me >> to write an entire wrapper for NSToolbar that will display native Mac >> toolbar lbuttons. >> >> While I've done some hacking and have added some bits to Tk's >> MacWindowStyle that allows one to simulate a native toolbar by applying >> Tk-compatible background colors to the underlying Cocoa window, wrapping >> NSToolbar is preferable because a) it's fully native and b) it can be >> set up as an extension. (The MacWindowStyle bits aren't available in the >> version of Tk that ships with Snow Leopard, and I can't ship my own >> build of Tk because of the private API issue.) >> >> Wish me luck! >> >> --Kevin >> > > > -- > Kevin Walzer > Code by Kevin > http://www.codebykevin.com > > ------------------------------------------------------------------------------ > The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: > Pinpoint memory and threading errors before they happen. > Find and fix more than 250 security defects in the development cycle. > Locate bottlenecks in serial and parallel code that limit performance. > http://p.sf.net/sfu/intel-dev2devfeb > _______________________________________________ > Tcl-mac mailing list > tc...@li... > https://lists.sourceforge.net/lists/listinfo/tcl-mac > |
From: Kevin W. <kw...@co...> - 2011-02-09 23:14:56
|
By way of comparison, here is the native toolbar: http://www.codebykevin.com/cocoa-toolbar.png and the emulated one: http://www.codebykevin.com/tk-unified-latest.png --K On 2/9/11 5:59 PM, Kevin Walzer wrote: > I've been doing a bit of hacking with Tk again, and I've discovered how > to enable a real NSToolbar in a Tk window via a Tk extension. Here's a > screenshot: > > http://www.codebykevin.com/cocoa-toolbar.png > > This is in the very early stages, but I think it will be feasible for me > to write an entire wrapper for NSToolbar that will display native Mac > toolbar lbuttons. > > While I've done some hacking and have added some bits to Tk's > MacWindowStyle that allows one to simulate a native toolbar by applying > Tk-compatible background colors to the underlying Cocoa window, wrapping > NSToolbar is preferable because a) it's fully native and b) it can be > set up as an extension. (The MacWindowStyle bits aren't available in the > version of Tk that ships with Snow Leopard, and I can't ship my own > build of Tk because of the private API issue.) > > Wish me luck! > > --Kevin > -- Kevin Walzer Code by Kevin http://www.codebykevin.com |
From: Kevin W. <kw...@co...> - 2011-02-09 22:59:19
|
I've been doing a bit of hacking with Tk again, and I've discovered how to enable a real NSToolbar in a Tk window via a Tk extension. Here's a screenshot: http://www.codebykevin.com/cocoa-toolbar.png This is in the very early stages, but I think it will be feasible for me to write an entire wrapper for NSToolbar that will display native Mac toolbar lbuttons. While I've done some hacking and have added some bits to Tk's MacWindowStyle that allows one to simulate a native toolbar by applying Tk-compatible background colors to the underlying Cocoa window, wrapping NSToolbar is preferable because a) it's fully native and b) it can be set up as an extension. (The MacWindowStyle bits aren't available in the version of Tk that ships with Snow Leopard, and I can't ship my own build of Tk because of the private API issue.) Wish me luck! --Kevin -- Kevin Walzer Code by Kevin http://www.codebykevin.com |
From: Barry S. <at...@me...> - 2011-02-06 16:22:31
|
All: I've been using this package as a notification system on Macs for several years now and it works well for our needs. I find Growl to slow start up times on my systems and do not use it, but wanted similar functionality for notifications in an App we were distributing. Also not everyone has Growl and this guarantees the available functionality in our App. To use it: package require tkgrowl ::tkgrowl::Growl "Title" "Message" This creates its own preferences folder and file and on first launch shows the preferences pane to users. You can call the preferences pane later by adding a menu item to your App that calls ::tkgrowl::Display_Preferences I currently do not have a Windows box to test against however it should theoretically work correctly. If anyone would like to test it and contribute any fixes or feedback that would be welcome. Standard BSD license applies and is included in the download. http://goodies.wizardacademypress.com/tkgrowl1.0.zip [ Around 8k zipped and 37k unzipped ] Thanks, Barry |
From: Kevin W. <kw...@co...> - 2011-02-05 02:15:12
|
Barry, On 2/4/11 8:46 PM, Barry Skidmore wrote: > I went ahead and made a .app out of this (and some other stuff). > Very nice. I think others will find it useful. I see you didn't include my cocoaprint package--good. I've been digging into it this week and have fixed a number of nasty bugs (printing extra copies, plus it is quite crash-prone on Snow Leopard). I'm going to release a new version as soon as SF restores its SVN and CVS servers. --Kevin -- Kevin Walzer Code by Kevin http://www.codebykevin.com |
From: Barry S. <at...@me...> - 2011-02-05 01:46:49
|
I went ahead and made a .app out of this (and some other stuff). This architecture is based on the 10.6 Wish and creates a .app package architecture to make it easy to alter and extend your app for the Mac platform. The app itself is 1.1 MB as a base package, your code adds weight. It links against the installed 10.6 TCL/Tk architecture. Any licenses from packages are included in the Contents/Resources/Frameworks directory. YMMV but the only limitations are those imposed by TCL and the included packages. Download: http://goodies.wizardacademypress.com/Application.zip Included Packages: appstorereceipt autoscroll macnotify tclAE2.0.5 Tclapplescript2.0 tclservices1.0 tkdnd2.1 tkdock1.0 tkmacicon1.1 windowlist This package should comply with all app store limitations as is. Your additions that may violate app store conditions not included :) Any code that can be strictly attributed to myself as part of this package is formally released into the public domain with no licensing restrictions or requirements whatsoever unless required by the included libraries. On Feb 4, 2011, at 8:51 AM, Kevin Walzer wrote: > Hi all, > > I've had some interest in the code I'm using to validate Mac App Store > receipts from Tcl/Tk apps and so I've posted the code here: > http://www.codebykevin.com/opensource/validatereceipt.zip. BSD license. > > Source code is included, but there is also a compiled 32/64-bit > Intel-only dylib included. > > It's very basic: it only implements receipt validation and does not > incorporate other anti-hack measures that have come to light since the > App Store went live. I based it on an earlier version of the code at > https://github.com/roddi/ValidateStoreReceipt. My reason for keeping it > very simple is that I don't have time to continually hack at it. > > Use is very easy: > > package require appstorereceipt > > appstorereceipt::validatereceipt > > If the receipt is not valid or not present, the app exits with a code of > 173 (per Apple's docs). > > I hope this is helpful! The package is also linked from my article on > preparing Tk apps for the App Store. > > --Kevin > > -- > Kevin Walzer > Code by Kevin > http://www.codebykevin.com > > ------------------------------------------------------------------------------ > The modern datacenter depends on network connectivity to access resources > and provide services. The best practices for maximizing a physical server's > connectivity to a physical network are well understood - see how these > rules translate into the virtual world? > http://p.sf.net/sfu/oracle-sfdevnlfb > _______________________________________________ > Tcl-mac mailing list > tc...@li... > https://lists.sourceforge.net/lists/listinfo/tcl-mac Thanks, Barry |
From: David Z. <zol...@lr...> - 2011-02-04 18:25:22
|
Le 4 févr. 2011 à 15:51, Kevin Walzer a écrit : > Hi all, > > I've had some interest in the code I'm using to validate Mac App Store > receipts from Tcl/Tk apps and so I've posted the code here: > http://www.codebykevin.com/opensource/validatereceipt.zip. BSD license. > > Source code is included, but there is also a compiled 32/64-bit > Intel-only dylib included. > > It's very basic: it only implements receipt validation and does not > incorporate other anti-hack measures that have come to light since the > App Store went live. I based it on an earlier version of the code at > https://github.com/roddi/ValidateStoreReceipt. My reason for keeping it > very simple is that I don't have time to continually hack at it. > > Use is very easy: > > package require appstorereceipt > > appstorereceipt::validatereceipt > > If the receipt is not valid or not present, the app exits with a code of > 173 (per Apple's docs). > > I hope this is helpful! The package is also linked from my article on > preparing Tk apps for the App Store. Great! Thanks a lot for this, Kevin! -- David Zolli kr...@kr... http://www.kroc.tk |
From: Kevin W. <kw...@co...> - 2011-02-04 14:51:27
|
Hi all, I've had some interest in the code I'm using to validate Mac App Store receipts from Tcl/Tk apps and so I've posted the code here: http://www.codebykevin.com/opensource/validatereceipt.zip. BSD license. Source code is included, but there is also a compiled 32/64-bit Intel-only dylib included. It's very basic: it only implements receipt validation and does not incorporate other anti-hack measures that have come to light since the App Store went live. I based it on an earlier version of the code at https://github.com/roddi/ValidateStoreReceipt. My reason for keeping it very simple is that I don't have time to continually hack at it. Use is very easy: package require appstorereceipt appstorereceipt::validatereceipt If the receipt is not valid or not present, the app exits with a code of 173 (per Apple's docs). I hope this is helpful! The package is also linked from my article on preparing Tk apps for the App Store. --Kevin -- Kevin Walzer Code by Kevin http://www.codebykevin.com |
From: Kevin W. <kw...@co...> - 2011-02-02 23:45:16
|
Hi all, I've posted an article that summarizes the steps I followed in successfully submitting my Tcl/Tk application to the Mac App Store. This article covers some of the same material that is available in Apple's docs (you have to pay the $99 developer fee to access those docs), but it focuses mainly on Tcl/Tk-specific issues and stuff I ran into. Here's the URL: http://www.codebykevin.com/opensource/app-store.html Hope this is of interest to others! --Kevin -- Kevin Walzer Code by Kevin http://www.codebykevin.com |
From: Jeff H. <je...@ac...> - 2011-01-27 18:39:28
|
I do suspect that it's just not been noticed in the Cocoa port to date. Please create a bug with the shortened snippet so we can track it (thanks for isolating to a small snippet). Jeff On 27/01/2011 3:22 AM, Wojciech Kocjan wrote: > I know it's strange to keep writing to yourself, but I did some more research. > > First of all, the following code reproduces the issue: > http://pastebin.com/ThUGFM4B > (it is now 30 lines and biggest part is logic I used when I first > tried>2 frames) > > Also, this also happens with Wish from OSX 10.6.2 so I am wondering if > noone has ever noticed this before on OSX? > > 2011/1/26 Wojciech Kocjan<woj...@ko...>: >> Jeff, >> >> I just doublechecked - you are indeed right. B4 is built using Carbon >> (hence 10.4+) and this is why it does not happen with B4. >> >> So, apart from moving back to Carbon (where biggest L&F loss are >> notebook tabs I think), is there any way to prevent the freezes? >> >> 2011/1/26 Jeff Hobbs<je...@ac...>: >>> FWIW, between AT 8.6b3 and 8.6b4 we switched from a universal x86-ppc 10.4+ >>> compile to x86-x64 10.5+ cocoa compile. >>> >>> On 26/01/2011 8:45 AM, Wojciech Kocjan wrote: >>>> >>>> One more update about this. >>>> >>>> I noticed 8.6.0 beta 3 has same problem, but 8.6.0 beta 4 does not. I >>>> looked at Tk ChangeLog between 2010-05-20 and 2010-10-07 and I found >>>> nothing related. There are some fixes in January, but I doubt they'd >>>> make it into beta 4 (unless they were posted earlier and made it into >>>> beta before being committed into Tk CVS HEAD). >>>> >>>> 2011/1/26 Wojciech Kocjan<woj...@ko...>: >>>>> >>>>> Hello, >>>>> >>>>> Recently I tried to upgrade to an OSX application to Tk 8.5.9 with >>>>> Cocoa (from https://github.com/das/tcltk/downloads) and I ran into a >>>>> strange issue. >>>>> >>>>> Occassionally the application freezes - especially when I use >>>>> grid/grid forget to switch between what is currently displayed. >>>>> >>>>> When building 8.5.9 without Cocoa (using Carbon) the issue does not >>>>> show up and GUI responses instantly. >>>>> I also tried ActiveTcl - 8.5.9.1 (which uses Cocoa) shows the same >>>>> issue. ActiveTcl 8.6.0.0b4 works fine and refreshes instantly. >>>>> >>>>> I have tried to create a minimal script to recreate it (perhaps it can >>>>> be even smaller, but I didn't try). Here it is: >>>>> http://pastebin.com/aTti34YR >>>>> >>>>> Clicking on Page 1 / Page 2 without moving mouse after clicking button >>>>> shows the issue. >>>>> Note that screen looks ok after a while or after mouse movements - >>>>> sometimes it takes 1-2 seconds to refresh, sometimes it waits for >>>>> mouse to move (in more complex GUIs). >>>>> >>>>> Does anyone have any idea what this might be and/or if there is a fix for >>>>> this? >>>>> >>>>> Perhaps this was also an issue in 8.6 and was fixed? I can try to port >>>>> the fix to 8.5. >>>>> >>>>> -- >>>>> WK |
From: Wojciech K. <woj...@ko...> - 2011-01-27 11:22:20
|
I know it's strange to keep writing to yourself, but I did some more research. First of all, the following code reproduces the issue: http://pastebin.com/ThUGFM4B (it is now 30 lines and biggest part is logic I used when I first tried >2 frames) Also, this also happens with Wish from OSX 10.6.2 so I am wondering if noone has ever noticed this before on OSX? 2011/1/26 Wojciech Kocjan <woj...@ko...>: > Jeff, > > I just doublechecked - you are indeed right. B4 is built using Carbon > (hence 10.4+) and this is why it does not happen with B4. > > So, apart from moving back to Carbon (where biggest L&F loss are > notebook tabs I think), is there any way to prevent the freezes? > > 2011/1/26 Jeff Hobbs <je...@ac...>: >> FWIW, between AT 8.6b3 and 8.6b4 we switched from a universal x86-ppc 10.4+ >> compile to x86-x64 10.5+ cocoa compile. >> >> On 26/01/2011 8:45 AM, Wojciech Kocjan wrote: >>> >>> One more update about this. >>> >>> I noticed 8.6.0 beta 3 has same problem, but 8.6.0 beta 4 does not. I >>> looked at Tk ChangeLog between 2010-05-20 and 2010-10-07 and I found >>> nothing related. There are some fixes in January, but I doubt they'd >>> make it into beta 4 (unless they were posted earlier and made it into >>> beta before being committed into Tk CVS HEAD). >>> >>> 2011/1/26 Wojciech Kocjan<woj...@ko...>: >>>> >>>> Hello, >>>> >>>> Recently I tried to upgrade to an OSX application to Tk 8.5.9 with >>>> Cocoa (from https://github.com/das/tcltk/downloads) and I ran into a >>>> strange issue. >>>> >>>> Occassionally the application freezes - especially when I use >>>> grid/grid forget to switch between what is currently displayed. >>>> >>>> When building 8.5.9 without Cocoa (using Carbon) the issue does not >>>> show up and GUI responses instantly. >>>> I also tried ActiveTcl - 8.5.9.1 (which uses Cocoa) shows the same >>>> issue. ActiveTcl 8.6.0.0b4 works fine and refreshes instantly. >>>> >>>> I have tried to create a minimal script to recreate it (perhaps it can >>>> be even smaller, but I didn't try). Here it is: >>>> http://pastebin.com/aTti34YR >>>> >>>> Clicking on Page 1 / Page 2 without moving mouse after clicking button >>>> shows the issue. >>>> Note that screen looks ok after a while or after mouse movements - >>>> sometimes it takes 1-2 seconds to refresh, sometimes it waits for >>>> mouse to move (in more complex GUIs). >>>> >>>> Does anyone have any idea what this might be and/or if there is a fix for >>>> this? >>>> >>>> Perhaps this was also an issue in 8.6 and was fixed? I can try to port >>>> the fix to 8.5. >>>> >>>> -- >>>> WK >> > |
From: Kevin W. <kw...@co...> - 2011-01-26 22:59:31
|
On 1/26/11 1:21 PM, Damon Courtney wrote: >> I'm pleased to report that I've successfully submitted one of my Tcl/Tk >> applications to the Mac App Store. Here's a blog entry on some of the >> details: >> >> http://bit.ly/hpH82t > > Excellent! I see from your blog that you had to revert back to the Carbon Tk to get it accepted, right? That's a real shame that Tk's Cocoa, which was (at least in part) paid for by Apple themselves can't even make it in. I didn't revert to Carbon--I linked to the Tk-Cocoa installed in /System/Library/Frameworks. That makes the private API calls allowed because it's installed by Apple and is part of the system. Not entirely logical, I know. :-) I'll say more about this later on my blog. > > I would also love a report (blog post or otherwise) about all the different pieces you're using in your apps now. In the last couple years you have become probably the most experienced Tcl/Tk application developer on the Mac, and you've done a lot of work to integrate with the native system. I know you've released some pieces yourself and contributed to others, but I would honestly love a run down of what you're using now and for what parts. Are there any pieces you use that you haven't already released? > > Just throwing that out there seeing as I'm sure you have TONS of time on your hands. 0-] > Well, actually, documenting this stuff in some format is a priority. I'm also going to begin work soon on documenting all the tk::mac magic bits and some of the Mac-specific C API bits, since so many folks have complained about it at the Tk tracker. --Kevin -- Kevin Walzer Code by Kevin http://www.codebykevin.com |
From: Damon C. <da...@tc...> - 2011-01-26 22:37:51
|
> I'm pleased to report that I've successfully submitted one of my Tcl/Tk > applications to the Mac App Store. Here's a blog entry on some of the > details: > > http://bit.ly/hpH82t Excellent! I see from your blog that you had to revert back to the Carbon Tk to get it accepted, right? That's a real shame that Tk's Cocoa, which was (at least in part) paid for by Apple themselves can't even make it in. > I'll be posting a more technical discussion of the mechanics of App > Store submission to the list and/or my blog in the very near future; > this will focus on both general issues and specific things I had to do > to get the Tcl/Tk app ready. (It was rejected twice, and the third time > was the charm.) > > Thanks to all your advice and encouragement, both on- and off-list. I'll > have more info to report soon. I would also love a report (blog post or otherwise) about all the different pieces you're using in your apps now. In the last couple years you have become probably the most experienced Tcl/Tk application developer on the Mac, and you've done a lot of work to integrate with the native system. I know you've released some pieces yourself and contributed to others, but I would honestly love a run down of what you're using now and for what parts. Are there any pieces you use that you haven't already released? Just throwing that out there seeing as I'm sure you have TONS of time on your hands. 0-] Thanks, D |
From: Jeff H. <je...@ac...> - 2011-01-26 22:37:44
|
FWIW, between AT 8.6b3 and 8.6b4 we switched from a universal x86-ppc 10.4+ compile to x86-x64 10.5+ cocoa compile. On 26/01/2011 8:45 AM, Wojciech Kocjan wrote: > One more update about this. > > I noticed 8.6.0 beta 3 has same problem, but 8.6.0 beta 4 does not. I > looked at Tk ChangeLog between 2010-05-20 and 2010-10-07 and I found > nothing related. There are some fixes in January, but I doubt they'd > make it into beta 4 (unless they were posted earlier and made it into > beta before being committed into Tk CVS HEAD). > > 2011/1/26 Wojciech Kocjan<woj...@ko...>: >> Hello, >> >> Recently I tried to upgrade to an OSX application to Tk 8.5.9 with >> Cocoa (from https://github.com/das/tcltk/downloads) and I ran into a >> strange issue. >> >> Occassionally the application freezes - especially when I use >> grid/grid forget to switch between what is currently displayed. >> >> When building 8.5.9 without Cocoa (using Carbon) the issue does not >> show up and GUI responses instantly. >> I also tried ActiveTcl - 8.5.9.1 (which uses Cocoa) shows the same >> issue. ActiveTcl 8.6.0.0b4 works fine and refreshes instantly. >> >> I have tried to create a minimal script to recreate it (perhaps it can >> be even smaller, but I didn't try). Here it is: >> http://pastebin.com/aTti34YR >> >> Clicking on Page 1 / Page 2 without moving mouse after clicking button >> shows the issue. >> Note that screen looks ok after a while or after mouse movements - >> sometimes it takes 1-2 seconds to refresh, sometimes it waits for >> mouse to move (in more complex GUIs). >> >> Does anyone have any idea what this might be and/or if there is a fix for this? >> >> Perhaps this was also an issue in 8.6 and was fixed? I can try to port >> the fix to 8.5. >> >> -- >> WK |
From: Wojciech K. <woj...@ko...> - 2011-01-26 22:37:03
|
Jeff, I just doublechecked - you are indeed right. B4 is built using Carbon (hence 10.4+) and this is why it does not happen with B4. So, apart from moving back to Carbon (where biggest L&F loss are notebook tabs I think), is there any way to prevent the freezes? 2011/1/26 Jeff Hobbs <je...@ac...>: > FWIW, between AT 8.6b3 and 8.6b4 we switched from a universal x86-ppc 10.4+ > compile to x86-x64 10.5+ cocoa compile. > > On 26/01/2011 8:45 AM, Wojciech Kocjan wrote: >> >> One more update about this. >> >> I noticed 8.6.0 beta 3 has same problem, but 8.6.0 beta 4 does not. I >> looked at Tk ChangeLog between 2010-05-20 and 2010-10-07 and I found >> nothing related. There are some fixes in January, but I doubt they'd >> make it into beta 4 (unless they were posted earlier and made it into >> beta before being committed into Tk CVS HEAD). >> >> 2011/1/26 Wojciech Kocjan<woj...@ko...>: >>> >>> Hello, >>> >>> Recently I tried to upgrade to an OSX application to Tk 8.5.9 with >>> Cocoa (from https://github.com/das/tcltk/downloads) and I ran into a >>> strange issue. >>> >>> Occassionally the application freezes - especially when I use >>> grid/grid forget to switch between what is currently displayed. >>> >>> When building 8.5.9 without Cocoa (using Carbon) the issue does not >>> show up and GUI responses instantly. >>> I also tried ActiveTcl - 8.5.9.1 (which uses Cocoa) shows the same >>> issue. ActiveTcl 8.6.0.0b4 works fine and refreshes instantly. >>> >>> I have tried to create a minimal script to recreate it (perhaps it can >>> be even smaller, but I didn't try). Here it is: >>> http://pastebin.com/aTti34YR >>> >>> Clicking on Page 1 / Page 2 without moving mouse after clicking button >>> shows the issue. >>> Note that screen looks ok after a while or after mouse movements - >>> sometimes it takes 1-2 seconds to refresh, sometimes it waits for >>> mouse to move (in more complex GUIs). >>> >>> Does anyone have any idea what this might be and/or if there is a fix for >>> this? >>> >>> Perhaps this was also an issue in 8.6 and was fixed? I can try to port >>> the fix to 8.5. >>> >>> -- >>> WK > |
From: Wojciech K. <woj...@ko...> - 2011-01-26 21:53:19
|
One more update about this. I noticed 8.6.0 beta 3 has same problem, but 8.6.0 beta 4 does not. I looked at Tk ChangeLog between 2010-05-20 and 2010-10-07 and I found nothing related. There are some fixes in January, but I doubt they'd make it into beta 4 (unless they were posted earlier and made it into beta before being committed into Tk CVS HEAD). 2011/1/26 Wojciech Kocjan <woj...@ko...>: > Hello, > > Recently I tried to upgrade to an OSX application to Tk 8.5.9 with > Cocoa (from https://github.com/das/tcltk/downloads) and I ran into a > strange issue. > > Occassionally the application freezes - especially when I use > grid/grid forget to switch between what is currently displayed. > > When building 8.5.9 without Cocoa (using Carbon) the issue does not > show up and GUI responses instantly. > I also tried ActiveTcl - 8.5.9.1 (which uses Cocoa) shows the same > issue. ActiveTcl 8.6.0.0b4 works fine and refreshes instantly. > > I have tried to create a minimal script to recreate it (perhaps it can > be even smaller, but I didn't try). Here it is: > http://pastebin.com/aTti34YR > > Clicking on Page 1 / Page 2 without moving mouse after clicking button > shows the issue. > Note that screen looks ok after a while or after mouse movements - > sometimes it takes 1-2 seconds to refresh, sometimes it waits for > mouse to move (in more complex GUIs). > > Does anyone have any idea what this might be and/or if there is a fix for this? > > Perhaps this was also an issue in 8.6 and was fixed? I can try to port > the fix to 8.5. > > -- > WK > |
From: Kevin W. <kw...@co...> - 2011-01-26 16:13:09
|
Hi all, I'm pleased to report that I've successfully submitted one of my Tcl/Tk applications to the Mac App Store. Here's a blog entry on some of the details: http://bit.ly/hpH82t I'll be posting a more technical discussion of the mechanics of App Store submission to the list and/or my blog in the very near future; this will focus on both general issues and specific things I had to do to get the Tcl/Tk app ready. (It was rejected twice, and the third time was the charm.) Thanks to all your advice and encouragement, both on- and off-list. I'll have more info to report soon. Thanks, Kevin -- Kevin Walzer Code by Kevin http://www.codebykevin.com |
From: Wojciech K. <woj...@ko...> - 2011-01-26 16:03:44
|
Hello, Recently I tried to upgrade to an OSX application to Tk 8.5.9 with Cocoa (from https://github.com/das/tcltk/downloads) and I ran into a strange issue. Occassionally the application freezes - especially when I use grid/grid forget to switch between what is currently displayed. When building 8.5.9 without Cocoa (using Carbon) the issue does not show up and GUI responses instantly. I also tried ActiveTcl - 8.5.9.1 (which uses Cocoa) shows the same issue. ActiveTcl 8.6.0.0b4 works fine and refreshes instantly. I have tried to create a minimal script to recreate it (perhaps it can be even smaller, but I didn't try). Here it is: http://pastebin.com/aTti34YR Clicking on Page 1 / Page 2 without moving mouse after clicking button shows the issue. Note that screen looks ok after a while or after mouse movements - sometimes it takes 1-2 seconds to refresh, sometimes it waits for mouse to move (in more complex GUIs). Does anyone have any idea what this might be and/or if there is a fix for this? Perhaps this was also an issue in 8.6 and was fixed? I can try to port the fix to 8.5. -- WK |
From: Ned D. <na...@ac...> - 2011-01-25 21:24:59
|
In article <4D3...@co...>, Kevin Walzer <kw...@co...> wrote: > It looks like you're mixing CVS HEAD of Tk (8.6) with a build of Tcl > 8.5. You need the same version of each. The best way to do this would be > to do an anon CVS checkout of HEAD of both Tcl and Tk 8.6. (Yes, the > official SCM of Tcl/Tk is still CVS.) Well, I don't think I'm doing that. Everything was from the das tcltk github tree. But, FWIW, I did a CVS checkout and I still see an install error with both tcl and tk: ../unix/install-sh: invalid option: -S make[4]: *** [install-binaries] Error 1 make[3]: *** [install-strip] Error 2 make[2]: *** [install-tk] Error 2 make[1]: *** [tk] Error 2 The unix/Makefile.in in both tcl and tk have this option: INSTALL_STRIP_LIBRARY = -S -x but the builtin install-sh script doesn't support that. Perhaps something needs to be overriden? > Also, remember, Tcl/Tk 8.6 isn't ready for production use--it's bleeding > edge. The Cocoa bits of Tk 8.6 work fine in a backported context to 8.5, > but that will have to wait until someone (Daniel Steffen or Jeff Hobbs) > does the appropriate merges. Torsten's patch won't make it back to 8.5 > until the Cocoa 8.5 backport is updated. Thanks for all the help. I am going to await an ActiveState update for OS X so, at this point, my interest in building Tcl/Tk is out of curiosity rather than a pressing need. -- Ned Deily, na...@ac... |
From: Kevin W. <kw...@co...> - 2011-01-25 15:53:31
|
On 1/24/11 7:08 PM, Ned Deily wrote: > In article<2FB...@ma...>, > Torsten Berg<re...@ma...> wrote: >>> Using the ActiveState Tcl/Tk Cocoa Tk 8.5.9 on OS X 10.6 results in >>> annoying repeated warning messages: >>> $ /usr/local/bin/wish8.5 >>> % 2011-01-24 12:20:42.061 Wish[6893:903] setCanCycle: is deprecated. >>> Please use setCollectionBehavior instead >>> 2011-01-24 12:20:42.068 Wish[6893:903] setCanCycle: is deprecated. >>> Please use setCollectionBehavior instead >> Interesting. I have never seen this message before on my 10.6 system. >>> So, for the ActiveState configuration (10.5 and 10.6, i386 and x86_64), >>> I assume this means that a change to the build configuration would >>> eliminate the messages on 10.6, so something like: >>> >>> export CFLAGS="-arch i386 -arch x86_64 \ >>> -isysroot /Developer/SDKs/MacOSX10.6.sdk" >>> export MACOSX_DEPLOYMENT_TARGET=10.5 >>> >> [...] >>> >>> Can anyone verify that building on 10.6 with a deployment target of 10.5 >>> does get rid of the warnings? Or can someone suggest the proper steps >>> to build successfully on 10.6? >> >> I normally build with settings something like that (using cvs HEAD): >> >> export CFLAGS='-mmacosx-version-min=10.5 -arch i386'make -C tcl${ver}/macosx >> >> make -C tk${ver}/macosx >> >> # never use ~ notation for directories ... >> make -C tcl${ver}/macosx install-embedded >> INSTALL_ROOT="$HOME/Tcl/distrib/mac-builds/${dest}/" >> make -C tk${ver}/macosx install-embedded >> INSTALL_ROOT="$HOME/Tcl/distrib/mac-builds/${dest}/" >> >> and this works fine without resulting in the message you mention. > > Thanks, Torsten. I guess I must be building from a different source. > I'm pulling the latest head from here: > > git clone https://github.com/das/tcltk.git > > The build fails first trying to make itcl: > > Configuring package 'itcl' > configure: loading cache ../../config.cache > checking for correct TEA configuration... ok (TEA 3.9) > checking whether ln -s works... yes > configure: error: cannot find install-sh or install.sh in tclconfig > [...]/tcltk/tcl/pkgs/itcl/tclconfig > make[2]: *** [configure-packages] Error 1 > make[1]: *** [build-tcl] Error 2 > make: *** [install-embedded-deploy] Error 2 > > and then: > > make install-tk \ > BUILD_STYLE=Deployment INSTALL_TARGET=install-strip EMBEDDED_BUILD=1 > INSTALL_BUILD=1 > [...] > ../tcltk/tk/unix/../generic/tk3d.c > In file included from ../tcltk/tk/unix/../generic/tkPort.h:23, > from ../tcltk/tk/unix/../generic/tkInt.h:21, > from ../tcltk/tk/unix/../generic/tk3d.c:16: > ../tcltk/tk/unix/../generic/tk.h:23:3: error: #error Tk 8.6 must be > compiled with tcl.h from Tcl 8.6 or better > In file included from ../tcltk/tk/unix/../generic/tkPort.h:23, > from ../tcltk/tk/unix/../generic/tkInt.h:21, > from ../tcltk/tk/unix/../generic/tk3d.c:16: > [followed by more compile errors] > Ned, It looks like you're mixing CVS HEAD of Tk (8.6) with a build of Tcl 8.5. You need the same version of each. The best way to do this would be to do an anon CVS checkout of HEAD of both Tcl and Tk 8.6. (Yes, the official SCM of Tcl/Tk is still CVS.) Also, remember, Tcl/Tk 8.6 isn't ready for production use--it's bleeding edge. The Cocoa bits of Tk 8.6 work fine in a backported context to 8.5, but that will have to wait until someone (Daniel Steffen or Jeff Hobbs) does the appropriate merges. Torsten's patch won't make it back to 8.5 until the Cocoa 8.5 backport is updated. --Kevin -- Kevin Walzer Code by Kevin http://www.codebykevin.com |
From: Torsten B. <re...@ma...> - 2011-01-25 15:42:47
|
> > Thanks, Torsten. I guess I must be building from a different source. > I'm pulling the latest head from here: > > git clone https://github.com/das/tcltk.git > I only build from the sources at sourceforge: http://tcl.sourceforge.net/ Torsten |
From: Ned D. <na...@ac...> - 2011-01-25 00:09:02
|
In article <2FB...@ma...>, Torsten Berg <re...@ma...> wrote: > > Using the ActiveState Tcl/Tk Cocoa Tk 8.5.9 on OS X 10.6 results in > > annoying repeated warning messages: > > $ /usr/local/bin/wish8.5 > > % 2011-01-24 12:20:42.061 Wish[6893:903] setCanCycle: is deprecated. > > Please use setCollectionBehavior instead > > 2011-01-24 12:20:42.068 Wish[6893:903] setCanCycle: is deprecated. > > Please use setCollectionBehavior instead > Interesting. I have never seen this message before on my 10.6 system. > > So, for the ActiveState configuration (10.5 and 10.6, i386 and x86_64), > > I assume this means that a change to the build configuration would > > eliminate the messages on 10.6, so something like: > > > > export CFLAGS="-arch i386 -arch x86_64 \ > > -isysroot /Developer/SDKs/MacOSX10.6.sdk" > > export MACOSX_DEPLOYMENT_TARGET=10.5 > > > [...] > > > > Can anyone verify that building on 10.6 with a deployment target of 10.5 > > does get rid of the warnings? Or can someone suggest the proper steps > > to build successfully on 10.6? > > I normally build with settings something like that (using cvs HEAD): > > export CFLAGS='-mmacosx-version-min=10.5 -arch i386'make -C tcl${ver}/macosx > > make -C tk${ver}/macosx > > # never use ~ notation for directories ... > make -C tcl${ver}/macosx install-embedded > INSTALL_ROOT="$HOME/Tcl/distrib/mac-builds/${dest}/" > make -C tk${ver}/macosx install-embedded > INSTALL_ROOT="$HOME/Tcl/distrib/mac-builds/${dest}/" > > and this works fine without resulting in the message you mention. Thanks, Torsten. I guess I must be building from a different source. I'm pulling the latest head from here: git clone https://github.com/das/tcltk.git The build fails first trying to make itcl: Configuring package 'itcl' configure: loading cache ../../config.cache checking for correct TEA configuration... ok (TEA 3.9) checking whether ln -s works... yes configure: error: cannot find install-sh or install.sh in tclconfig [...]/tcltk/tcl/pkgs/itcl/tclconfig make[2]: *** [configure-packages] Error 1 make[1]: *** [build-tcl] Error 2 make: *** [install-embedded-deploy] Error 2 and then: make install-tk \ BUILD_STYLE=Deployment INSTALL_TARGET=install-strip EMBEDDED_BUILD=1 INSTALL_BUILD=1 [...] ../tcltk/tk/unix/../generic/tk3d.c In file included from ../tcltk/tk/unix/../generic/tkPort.h:23, from ../tcltk/tk/unix/../generic/tkInt.h:21, from ../tcltk/tk/unix/../generic/tk3d.c:16: ../tcltk/tk/unix/../generic/tk.h:23:3: error: #error Tk 8.6 must be compiled with tcl.h from Tcl 8.6 or better In file included from ../tcltk/tk/unix/../generic/tkPort.h:23, from ../tcltk/tk/unix/../generic/tkInt.h:21, from ../tcltk/tk/unix/../generic/tk3d.c:16: [followed by more compile errors] -- Ned Deily, na...@ac... |
From: Torsten B. <re...@ma...> - 2011-01-24 22:13:41
|
Hi, > Using the ActiveState Tcl/Tk Cocoa Tk 8.5.9 on OS X 10.6 results in > annoying repeated warning messages: > > $ /usr/local/bin/wish8.5 > % 2011-01-24 12:20:42.061 Wish[6893:903] setCanCycle: is deprecated. > Please use setCollectionBehavior instead > 2011-01-24 12:20:42.068 Wish[6893:903] setCanCycle: is deprecated. > Please use setCollectionBehavior instead > Interesting. I have never seen this message before on my 10.6 system. > > So, for the ActiveState configuration (10.5 and 10.6, i386 and x86_64), > I assume this means that a change to the build configuration would > eliminate the messages on 10.6, so something like: > > export CFLAGS="-arch i386 -arch x86_64 \ > -isysroot /Developer/SDKs/MacOSX10.6.sdk" > export MACOSX_DEPLOYMENT_TARGET=10.5 > [...] > > Can anyone verify that building on 10.6 with a deployment target of 10.5 > does get rid of the warnings? Or can someone suggest the proper steps > to build successfully on 10.6? I normally build with settings something like that (using cvs HEAD): export CFLAGS='-mmacosx-version-min=10.5 -arch i386'make -C tcl${ver}/macosx make -C tk${ver}/macosx # never use ~ notation for directories ... make -C tcl${ver}/macosx install-embedded INSTALL_ROOT="$HOME/Tcl/distrib/mac-builds/${dest}/" make -C tk${ver}/macosx install-embedded INSTALL_ROOT="$HOME/Tcl/distrib/mac-builds/${dest}/" and this works fine without resulting in the message you mention. Regards, Torsten |
From: Ned D. <na...@ac...> - 2011-01-24 20:33:39
|
Using the ActiveState Tcl/Tk Cocoa Tk 8.5.9 on OS X 10.6 results in annoying repeated warning messages: $ /usr/local/bin/wish8.5 % 2011-01-24 12:20:42.061 Wish[6893:903] setCanCycle: is deprecated. Please use setCollectionBehavior instead 2011-01-24 12:20:42.068 Wish[6893:903] setCanCycle: is deprecated. Please use setCollectionBehavior instead ... Similar messages are seen with Python's IDLE, my concern. The topic has come up before and in this message, Daniel suggests the solution: http://article.gmane.org/gmane.comp.lang.tcl.mac/6524/ So, for the ActiveState configuration (10.5 and 10.6, i386 and x86_64), I assume this means that a change to the build configuration would eliminate the messages on 10.6, so something like: export CFLAGS="-arch i386 -arch x86_64 \ -isysroot /Developer/SDKs/MacOSX10.6.sdk" export MACOSX_DEPLOYMENT_TARGET=10.5 To test that, I've tried building a Cocoa Tcl/Tk on 10.6 using the current git checkout from http://github.com/das/tcltk/tree/de-carbon-8-5 but I run into various errors in the build using the above exports with: ver="" make -C tcl${ver}/macosx make -C tcl${ver}/macosx install-deploy Can anyone verify that building on 10.6 with a deployment target of 10.5 does get rid of the warnings? Or can someone suggest the proper steps to build successfully on 10.6? It would be great if the messages could be eliminated from the next ActiveState release. -- Ned Deily, na...@ac... |
From: Ned D. <na...@ac...> - 2011-01-24 20:19:18
|
In article <4D3...@ac...>, Jeff Hobbs <je...@ac...> wrote: > On 23/01/2011 10:56 AM, Ned Deily wrote: > > In > > article<D2CFDFB0-A7CB-4468-8660-BF346BB49120-dKc1CtW5WJqELgA04lAiVw@public.g > > mane.org>, > > Torsten Berg<re...@ma...> wrote: > >> I just uploaded a pathced version of tkMacOSXKeyEvent.c to the bug at SF. > > > > Thanks, Torsten. And, thanks, Kevin, for looking into this. I don't > > really want to get us into the business of building and shipping Tk with > > python.org installers but, if an official patch prevents this nasty > > crash and doesn't cause any other regressions, it might be worth it. > > I'll make sure that we have an ActiveTcl release updated with this patch > prior to Feb 12. Note that we are using the 8.5-decarbon branch created > by Daniel Steffen, which is not the "official" 8.5 source branch (that > one still has the carbon sources). Thus, no need for an official 8.5 > release related to this, and I suspect most people aren't using Tkinter > and 8.6 at this time. Jeff, That's great! Thanks so much to you and Kevin and Torsten. If there's anything we can do to test it, let me know. While I have your combined attention, there's one other non-critical issue with Cocoa Tk 8.5 that could perhaps be addressed, that is the "setCanCycle: is deprecated" running on 10.6. I'll start a new thread about that. -- Ned Deily, na...@ac... |