- Dillo-dev - lists.mailman3.com

FreeHG performance

by onepoint＠starurchin.org

On Thu, Feb 05, 2009 at 05:52:40AM -0700, Scott Scriven wrote: > Did anyone ever look at launchpad? > > https://launchpad.net/+tour > > It's a "batteries included" type of project infrastructure host, > feature-rich, easy, reliable, and free. It could provide hosting > for almost everything the project needs and provide plenty of > room to grow. Looks good, but I think Dillo would have to migrate away from Mercurial to use it. They use bzr, and although they integrate with some other VCSs, mercurial isn't one of them. Regards, Jeremy Henty

Feb. 5, 2009

FreeHG performance

by dillo-dev＠toykeeper.net

* Jeremy Henty <onepoint(a)starurchin.org> wrote: > On Tue, Feb 03, 2009 at 01:47:48PM +0100, Hofmann Johannes wrote: > > I doubt that setting up a savannah project would take very long. > Really? I just had a look and there's a long checklist. ... Did anyone ever look at launchpad? https://launchpad.net/+tour It's a "batteries included" type of project infrastructure host, feature-rich, easy, reliable, and free. It could provide hosting for almost everything the project needs and provide plenty of room to grow. The tour gives a pretty good overview... it may be a lot to absorb all at once, but it's worth a look. -- Scott

Feb. 5, 2009

Re: view stylesheets from page menu

by Johannes.Hofmann＠gmx.de

On Thu, Feb 05, 2009 at 08:19:03AM -0300, Jorge Arellano Cid wrote: > On Tue, Feb 03, 2009 at 08:00:12PM +0100, Hofmann Johannes wrote: > > Hi corvid, > > > > On Tue, Feb 03, 2009 at 05:13:27PM +0000, corvid wrote: > > > > > > (a rough quick patch just to get feedback) > > > I saw that Jorge added cssUrls, which provided an opportunity... > > > > Very nice! Just one small thing. The stylesheets entry does not get > > disabled on non-CSS pages here. Extending the check like this: > > > > + if (cssUrls && cssUrls->size () > 0) { > > > > helps. > > To me: It looks more like a developers feature than one for user level. Well, not more than "View page source". I think it's important and in the spirit of dillo to make the web as transparent for the user as possible. > > > > BTW: Viewing stylesheets in a normal browser window is really fast > > compared to the fltk text viewer we use for "View page source"! > > Yes, since a long time I've wanted to find some time to turn > "View page source" into a dpi. > > Maybe after the close-tab button I can give it a hand. > Yes, that would be great. Cheers, Johannes > -- > Cheers > Jorge.-

Feb. 5, 2009

FreeHG performance

by jcid＠dillo.org

On Wed, Feb 04, 2009 at 08:31:06PM +0100, Andreas Kemnade wrote: > Hi, > > On Fri, 30 Jan 2009 17:24:56 +0100 > Andreas Kemnade <akemnade(a)tzi.de> wrote: > > > Hi Jorge, > > > > On Fri, 30 Jan 2009 12:22:46 -0300 > > Jorge Arellano Cid <jcid(a)dillo.org> wrote: > > > > > On Fri, Jan 30, 2009 at 01:21:47PM +0000, Jeremy Henty wrote: > > > > > > > > Am I the only one getting very poor performance from FreeHG ? Quite > > > > often I get a 404 Not Found and right now it's hanging. Everyone > > > > confirms that Mercurial itself is really fast and I've seen other > > > > reports criticising FreeHG's performance. Perhaps we should think > > > > about moving host? It's not always easy to find time to devote to > > > > Dillo so it's rather aggravating when that time is wasted just waiting > > > > for a connection to the server. I'm committed enough to grit my teeth > > > > and wait but I can see it putting off would-be contributors. > > > > > > > > Just a thought, > > > > > > Yes, I've also had trouble with it, and when normal it's slower > > > than our Wearlab's CVS. > > > > > > A couple of days ago (while developoing/reviewing patches) I > > > digged a bit on it and it seems that bitbucket is much faster. We > > > can try bitbucket while we find a long term solution (e.g. FSF's > > > savannah is offering Mercurial and Git repos, but the applying > > > process may take some time). > > > > > > At some point in time we should move out from the Wearlab, as > > > it may be closed soon. Currently we need webserver/CGI/python > > > shell accounts and a DCVS (Hg or Git). CGI is quite useful, but > > > we may use another bugtracker too. > > > > > The server which actually hosts the dillo debian and openmoko > > packages should be an alternative. > > At least performance and bandwidth should be no problem. Maybe moving > > hg there is a starting point. > > > ok, try > hg clone http://hg.dillo.org/dillo > > just add entries for hg.dillo.org in your /etc/hosts to point to that server. > 85.214.105.140 hg.dillo.org It works very well! > push access (with ssl) is not set up yet. > When it is done, of course the dns can be changed so that > hg.dillo.org points to the right server. Good. Just let me know when it's done. -- Cheers Jorge.-

Feb. 5, 2009

FreeHG performance

by onepoint＠starurchin.org

On Wed, Feb 04, 2009 at 08:31:06PM +0100, Andreas Kemnade wrote: > ok, try hg clone http://hg.dillo.org/dillo Now *that* was nice and fast! Thanks! Jeremy Henty

Feb. 4, 2009

FreeHG performance

by akemnade＠tzi.de

Hi, On Fri, 30 Jan 2009 17:24:56 +0100 Andreas Kemnade <akemnade(a)tzi.de> wrote: > Hi Jorge, > > On Fri, 30 Jan 2009 12:22:46 -0300 > Jorge Arellano Cid <jcid(a)dillo.org> wrote: > > > On Fri, Jan 30, 2009 at 01:21:47PM +0000, Jeremy Henty wrote: > > > > > > Am I the only one getting very poor performance from FreeHG ? Quite > > > often I get a 404 Not Found and right now it's hanging. Everyone > > > confirms that Mercurial itself is really fast and I've seen other > > > reports criticising FreeHG's performance. Perhaps we should think > > > about moving host? It's not always easy to find time to devote to > > > Dillo so it's rather aggravating when that time is wasted just waiting > > > for a connection to the server. I'm committed enough to grit my teeth > > > and wait but I can see it putting off would-be contributors. > > > > > > Just a thought, > > > > Yes, I've also had trouble with it, and when normal it's slower > > than our Wearlab's CVS. > > > > A couple of days ago (while developoing/reviewing patches) I > > digged a bit on it and it seems that bitbucket is much faster. We > > can try bitbucket while we find a long term solution (e.g. FSF's > > savannah is offering Mercurial and Git repos, but the applying > > process may take some time). > > > > At some point in time we should move out from the Wearlab, as > > it may be closed soon. Currently we need webserver/CGI/python > > shell accounts and a DCVS (Hg or Git). CGI is quite useful, but > > we may use another bugtracker too. > > > The server which actually hosts the dillo debian and openmoko > packages should be an alternative. > At least performance and bandwidth should be no problem. Maybe moving > hg there is a starting point. > ok, try hg clone http://hg.dillo.org/dillo just add entries for hg.dillo.org in your /etc/hosts to point to that server. 85.214.105.140 hg.dillo.org or 2001:6f8:900:71b::2 hg.dillo.org push access (with ssl) is not set up yet. When it is done, of course the dns can be changed so that hg.dillo.org points to the right server. Greetings Andreas Kemnade

Feb. 4, 2009

Leaking image structures

by onepoint＠starurchin.org

One of the biggest leaks that valgrind reports is image structures. I'm attaching logs that show memory being allocated by a_Png_new() and cousins and not being freed. There are also frequent leaks of memory allocated by the image libraries from within the various image callbacks, but I suspect these are just another effect of the Dillo image structures not being freed. Unfortunately I'm having a hard time tracking these leaks down. They regularly appear after long browsing sessions, but I haven't found any single page that reproduces them. I suspect they are caused by some race condition inside the CCC rather than the HTML parser. I am browsing with image loading enabled, and if I understand the code correctly that means that the Web structures in the CCC own all the references to image structures. The fact that image structures are reference counted makes it hard to know when the structures should be freed anyway. Indeed, does Dillo even ensure that image structures are freed? Maybe this is just a symptom of incomplete cleanup at exit rather than a true leak? I've looked over the code and can't see any obvious leaks so I'm rather stuck. Any tips for tracking this down? Even if it's not a true leak it would be good to have a fix that cleaned up the valgrind logs. Regards, Jeremy Henty

Feb. 4, 2009

[patch] Search Previous

by jcid＠dillo.org

On Wed, Feb 04, 2009 at 08:51:26AM +0100, Hofmann Johannes wrote: > On Tue, Feb 03, 2009 at 11:54:56PM +0000, Jo?o Ricardo Louren?o wrote: > > Ter, 2009-02-03 ?s 21:39 +0100, Hofmann Johannes escreveu: > > > Hello, > > > > > > On Tue, Feb 03, 2009 at 07:48:18PM +0000, Jo?o Ricardo Louren?o wrote: > > > > Greetings. > > > > > > > > I 'spotted' a TODO on searching 'backwards' and decided to have a go at > > > > it. > > > > > > Great. > > > > > > > > > > > Basically, I implemented the Search Previous feature. > > > > > > > > I must say I have difficulties understanding the code behind > > > > FindtextState::search0(), so I did not base my code on it at all. Sure > > > > it could possibly be optimized, but it seems to do the job for now, I > > > > haven't found any bugs, if you find any, please do say. > > > > > > FindtextState implements the Knuth-Morris-Pratt algorithm > > > http://en.wikipedia.org/wiki/Knuth%E2%80%93Morris%E2%80%93Pratt_algorithm > > > for efficient searching. It is pretty clever but a bit tricky. > > > I see some issues in your implementation. E.g. try to search for > > > 'ab' in a string 'abababa'. > > > Even more tricky is searching for ababb in abababb, i.e. the prefix > > > matches, but once it no longer does you would need to backtrack. > > > The easiest but inefficient way to implement searching > > > is doing a strcmp() at every text position. The Knuth-Morris-Pratt > > > algorithm avoids that. > > > I think it would be best to adapt search0 somehow to allow backward > > > searching - it's more or less the same thing as forward searching > > > anyway. Some ugly edge cases when switching between forward and > > > backward maybe :-) > > > > > > Thanks for working on that, > > > Johannes > > > > > > > > > > > Additionally, I did not set any hotkeys for the search-previous action, > > > > maybe something like SHIFT+RETURN, since RETURN handles searching > > > > forwards? > > > > > > > > Also, I did not add any tests, and I only altered a line to make them > > > > compilable, but I might add some tests soon. > > > > > > > > Thanks, > > > > > > > > Jo?o > > > > Thanks for your answer. > > > > OK, I am still trying to 'go deep' with the KMP algorithm, but I have > > got (provisory) working search0Backwards() replacement. How efficient do > > you think it is to handle the word by reversing it? Rephrasing that, do > > you think there is a 'really' better way of searching backwards, than > > reversing the word and searching backwards? As far as code-changes it > > takes very few lines and I have a working example of it. However, the > > extra reversing work might hit us hard. > > Reversing the word (needle) we are searching for is no problem. It's supposed > to be rather small and we only need to reverse it once. > The point is to avoid the backtracking. So that every letter in the > text (haystack) is only visited once. If reversing the needle is all it takes, go ahead. If the code is simple and it takes twice the time a forward search, go ahead too. Backward search is much less used than forward search and it's better to have it than not, and unless the delay is annoying there's no compelling need to optimize. -- Cheers Jorge.-

Feb. 4, 2009

[patch] Search Previous

by Johannes.Hofmann＠gmx.de

On Tue, Feb 03, 2009 at 11:54:56PM +0000, Jo?o Ricardo Louren?o wrote: > Ter, 2009-02-03 ?s 21:39 +0100, Hofmann Johannes escreveu: > > Hello, > > > > On Tue, Feb 03, 2009 at 07:48:18PM +0000, Jo?o Ricardo Louren?o wrote: > > > Greetings. > > > > > > I 'spotted' a TODO on searching 'backwards' and decided to have a go at > > > it. > > > > Great. > > > > > > > > Basically, I implemented the Search Previous feature. > > > > > > I must say I have difficulties understanding the code behind > > > FindtextState::search0(), so I did not base my code on it at all. Sure > > > it could possibly be optimized, but it seems to do the job for now, I > > > haven't found any bugs, if you find any, please do say. > > > > FindtextState implements the Knuth-Morris-Pratt algorithm > > http://en.wikipedia.org/wiki/Knuth%E2%80%93Morris%E2%80%93Pratt_algorithm > > for efficient searching. It is pretty clever but a bit tricky. > > I see some issues in your implementation. E.g. try to search for > > 'ab' in a string 'abababa'. > > Even more tricky is searching for ababb in abababb, i.e. the prefix > > matches, but once it no longer does you would need to backtrack. > > The easiest but inefficient way to implement searching > > is doing a strcmp() at every text position. The Knuth-Morris-Pratt > > algorithm avoids that. > > I think it would be best to adapt search0 somehow to allow backward > > searching - it's more or less the same thing as forward searching > > anyway. Some ugly edge cases when switching between forward and > > backward maybe :-) > > > > Thanks for working on that, > > Johannes > > > > > > > > Additionally, I did not set any hotkeys for the search-previous action, > > > maybe something like SHIFT+RETURN, since RETURN handles searching > > > forwards? > > > > > > Also, I did not add any tests, and I only altered a line to make them > > > compilable, but I might add some tests soon. > > > > > > Thanks, > > > > > > Jo?o > > Thanks for your answer. > > OK, I am still trying to 'go deep' with the KMP algorithm, but I have > got (provisory) working search0Backwards() replacement. How efficient do > you think it is to handle the word by reversing it? Rephrasing that, do > you think there is a 'really' better way of searching backwards, than > reversing the word and searching backwards? As far as code-changes it > takes very few lines and I have a working example of it. However, the > extra reversing work might hit us hard. Reversing the word (needle) we are searching for is no problem. It's supposed to be rather small and we only need to reverse it once. The point is to avoid the backtracking. So that every letter in the text (haystack) is only visited once. Regards, Johannes > > Thanks, > > Jo?o > > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

Feb. 4, 2009

[patch] Search Previous

by jorl17.8＠gmail.com

Ter, 2009-02-03 ?s 21:39 +0100, Hofmann Johannes escreveu: > Hello, > > On Tue, Feb 03, 2009 at 07:48:18PM +0000, Jo?o Ricardo Louren?o wrote: > > Greetings. > > > > I 'spotted' a TODO on searching 'backwards' and decided to have a go at > > it. > > Great. > > > > > Basically, I implemented the Search Previous feature. > > > > I must say I have difficulties understanding the code behind > > FindtextState::search0(), so I did not base my code on it at all. Sure > > it could possibly be optimized, but it seems to do the job for now, I > > haven't found any bugs, if you find any, please do say. > > FindtextState implements the Knuth-Morris-Pratt algorithm > http://en.wikipedia.org/wiki/Knuth%E2%80%93Morris%E2%80%93Pratt_algorithm > for efficient searching. It is pretty clever but a bit tricky. > I see some issues in your implementation. E.g. try to search for > 'ab' in a string 'abababa'. > Even more tricky is searching for ababb in abababb, i.e. the prefix > matches, but once it no longer does you would need to backtrack. > The easiest but inefficient way to implement searching > is doing a strcmp() at every text position. The Knuth-Morris-Pratt > algorithm avoids that. > I think it would be best to adapt search0 somehow to allow backward > searching - it's more or less the same thing as forward searching > anyway. Some ugly edge cases when switching between forward and > backward maybe :-) > > Thanks for working on that, > Johannes > > > > > Additionally, I did not set any hotkeys for the search-previous action, > > maybe something like SHIFT+RETURN, since RETURN handles searching > > forwards? > > > > Also, I did not add any tests, and I only altered a line to make them > > compilable, but I might add some tests soon. > > > > Thanks, > > > > Jo?o Thanks for your answer. OK, I am still trying to 'go deep' with the KMP algorithm, but I have got (provisory) working search0Backwards() replacement. How efficient do you think it is to handle the word by reversing it? Rephrasing that, do you think there is a 'really' better way of searching backwards, than reversing the word and searching backwards? As far as code-changes it takes very few lines and I have a working example of it. However, the extra reversing work might hit us hard. Thanks, Jo?o

Feb. 4, 2009