- Dillo-dev - lists.mailman3.com

[RFC] Find as you type, statusbar instead of popups, etc
by Frank de Lange Sept. 11, 2003

Sept. 11, 2003

Hi'all, My current tab/frame patch (which will probably be released AFTER the 20th of september, as I am going abroad for a while) contains a find_as_you_type implementation (like Mozilla, search results are updated interactively). I also included an option to use the status bar and location bar instead of popup windows for search and open_url (popups are quite often not worth the hassle...). User controls for find (and find_as_you_type) are the same as the normal find dialog ("/" or CTRL-F to start find, ESC to close, enter to search/search next), but find now retains the last search string between searches. Code size impact is minimal, in case of a popup_less version code is actually quite a bit less complex (and a bit smaller) than the current CVS implementation. Is this something you'd like to see in Dillo, or should I keep these to myself? On the frameset front, control over margins in (i)frames has been improved (no more extraneous margin space in iframe). Dillo (when configured for frameset support) now always renders frameset - even when body content is placed outside the frameset. Frame resize support has been improved a bit. These changes introduce two new preferences (popup_dialogs and find_as_you_type). Both have defaults which reflect unpatched Dillo's behaviour. Please let me know if this sounds like an improvement (or the opposite) to you. Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

3 3

[PATCHES] GTK1 focus problem, solution
by Frank de Lange Sept. 10, 2003

Sept. 10, 2003

Jorge, Sebastian, world, In a previous message I described a focus problem with Dillo (focus is 'trapped' in invisible widget). I have traced the problem to the GtkToolbar class. GtkToolbar is a GtkContainer, so it always gets focus even when GTK_CAN_FOCUS is unset. Hoewever, it redefines the focus signal function to NULL, thereby trapping focus inside the widget. To get normal focus behaviour, the GtkToolbar class's focus function needs to be changed to something like { return FALSE; }. I have patched interface.c to this extent. Patch is attached to this message. To get a complete, two-way tab focus chain (ability to tab around all focusable widgets in Dillo, forward and backward) my focus patch for dw_gtk_scrolled_frame is needed, otherwise dw_gtk_scrolled_frame will trap focus inside itself. You can find the patches on my patch site: http://www.geocities.com/ikbenfrank/dillo-200309101437-interface_focus.patc… (size: 660 bytes) http://www.geocities.com/ikbenfrank/dillo-200309101116-focus_chain.patch.gz (size: 910 bytes) Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

1 0

[PATCHES] hand_cursor and image_popup, some remarks
by Frank de Lange Sept. 10, 2003

Sept. 10, 2003

[ Jorge, your mailserver is not healthy... I tried to send this 3 times without success, so it goes to the list instead... ] On Mon, 8 Sep 2003 11:13:32 -0400 (CLT) Jorge Arellano Cid <jcid(a)softhome.org> wrote: > You're right. In that case it causes more troubles than help. > Please send me the patch to commit it to CVS. It is part of the tab/frame patch... It uses the document abstraction. In non-patched Dillo, it would be something like: if(bw->nav_stack_ptr != -1) gtk_widget_grab_focus(GTK_BIN(bw->docwin)->child); else gtk_widget_grab_focus(bw->location); Question is, how much sense does this make in a single-document window? When the window is created, the navigation stack is empty, so focus will always be on the location bar... By the way, there are some problems with 'hidden focus' in Dw. These problems only become directly noticeable in the tab/frame patched version, but they are present in CVS Dillo as well. To see the problem: 1. start dillo on a longish directory listing or a long document (so that the vertical scroll bar appears) 2. use up and down arrow keys to scroll the document. This works in unpatched Dillo. In the (current) patched version, it only works AFTER pressing the UP key once... 3. press SHIFT-TAB twice. Focus is now set on... something. Something invisible. There is no visible focus cue 4. press DOWN key. Nothing happens, no matter how often you press it... Same for TAB, ENTER, RIGHT, ... 5. press UP or SHIFT-TAB or LEFT key. Focus is now on the document, and scrolling keys work as they should. The problem does not show (in patched or unpatched Dillo) when in fullscreen mode. I've searched for the cause of this problem, but have not yet found it. > Clashing with other apps. is almost unavoidable (unless we > clone), but we have chosen to rely over an underlying cognitive > model for our UI. It has been very succesful, and praised! > > Now, as a matter of serendipity :), the bouncing keyboard may > have shown us both a better way. If we make Ctrl-L focus the > location bar (as Ctrl-U does now) and a second Ctrl-L to clear > it, that may be a good choice. Well, Ctrl-U is generally used for 'clear entry' in GTK1 (not in GTK2 though), so I think it is best to use it for that purpose in Dillo as well. For GTK2, following the toolkit default bindings is the logical thing to do. Even better is to allow users to decide for themselves which bindings they want. GTK2 does not allow this by default, but it can be re-enabled. No matter how well thought out a model is, the user should *always* be able to override the designer's defaults. Even if that user messes up a perfectly thought out cognitive model. That is also why I make just about everything user-selectable using preferences with sane default values. So, one of my next goals will probably be to implement user-selectable accelerators in Dillo... Based on GTK's default mechanisms for doing such a thing. > Of course, a more orthogonal way to do it, is the same > functionality, but inside the location popup. But please > experiment with the above suggested idea, and tell me your > findings. Generally I find that popup menu's are more a hinder than a help. This is more of a problem with window managers which do not correctly manage them than with the basic concept of popups, but as the state of the art currently stands I try to avoid them as much as possible. I will therefore try to implement the functionality of Dillo's transient popups (find and open url) in the main window, the former probably in the status bar and the latter in the location bar. And, of course, make a preference to allow the user to choose which version to use - popup or main window. As for using CTRL-L for clearing the location bar, I do not think this is a good idea. But, it is not for me to say it is a bad idea either. > PS: If you send me the requested patches, chances of fast commit > are very high Attached two patches, one for hand_cursor, one for right_click on image. Some remarks: The 'image save as' function in the image popup works, but is dependent on the dpi for the actual save function. That dpi (or, rather, wget) does not know how to handle some link types (eg. file: links), so it will fail on those. The hand_cursor patch works fine, but input_images do NOT work as stated in my previous message. Reverse the patch to dw_button.c to get them back to live... Maybe the image popup should be a submenu on other popups? See the frame/frameset popups (in the tab/frame patch) for an example of what I mean. That way, the image can also be accessed in links (and, maybe, in input images, even those currently do not react to RMB-clicks for obvious reason...) Please be aware of the fact that I will have to keep on adding these patches to the tab/frame patch while the document abstraction has not been committed to CVS... So I will have to patch my own patches, not something I want to keep on doing too long... Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

2 2

FW: [Dillo-dev] How do your clear the dillo cache
by Ronald W. Henderson Sept. 9, 2003

Sept. 9, 2003

Folks: Let me be more accurate with my problem: The graphic images (which are *.png files) are the ones which are cached. If I hit the reload botton the html code updates ok but the graphics don't. I need to start and stop Dillo for the graphic images to be updated. Folks: How do you clear the Dillo cache? I am using dillo (0.7.3) to view an updated local html file: "file:xxx.html". I have to stop and restart dillo to see the updates. The "Reload" Button or open file menu option just keeps loading the cached pages... Thanks.. _______________________________________________ Dillo-dev mailing list Dillo-dev(a)lists.auriga.wearlab.de http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

2 1

Two small things that need implementation...
by Jorge Arellano Cid Sept. 9, 2003

Sept. 9, 2003

Hi, These things have waited for a long time unimplemented. Both don't require much code, but some knowledge: 1.- A right-click popup for images (we have all the code and functions of it, what we need is to hook an empty popup to right-click over images). 2.- Input images work, but there's a detail lacking. The arrow- shaped cursor should change to a hand-shaped one (as with links). Anyone? Cheers Jorge.-

3 3

Feature request - Open tab in background
by Colin Michael Yates Sept. 8, 2003

Sept. 8, 2003

Hi, Is it possible to open a tab in the background, instead of it always getting focus? I generally go to a forum (for example), and open the messages that I want to read in a seperate tab, when I have selected all the messages I want to read, I then go through them one by one. Is this feature already implemented, or does anybody with more C coding skill then me want to implement it :) Cheers, Colin

2 1

How do your clear the dillo cache
by Ronald W. Henderson Sept. 7, 2003

Sept. 7, 2003

Folks: How do you clear the Dillo cache? I am using dillo (0.7.3) to view an updated local html file: "file:xxx.html". I have to stop and restart dillo to see the updates. The "Reload" Button or open file menu option just keeps loading the cached pages... Thanks..

1 0

Check for unterminated quoted strings
by Andreas Kemnade Sept. 6, 2003

Sept. 6, 2003

I found this tag on pages at www.reichelt.de: <A HREF="Info.html?P1mqhKwQAQsAAHVmTEY+Suche+0+50+LM%20324%20D%20SMD+lm324" onmouseover="return overlib('<TABLE><TR><TD><font face=\'Verdana, Arial, Helvetica, sans-serif\' size=1 color=\'#FFFFFF\'></font></TD></TR></TABLE>',CAPTION,'LM 324 D SMD');" onMouseout="return nd();"> It looks like correct html to me or am I wrong? But dillo does not render it correctly. It is rendered like this: <A HREF=... onmouseover="...TABLE"><TR><TD>... So the check for unterminated quoted strings confuses the table handling. With a small change to html.c, it will be rendered correctly: /* Skip over quoted string */ buf_index++; buf_index += strcspn(buf + buf_index, - (ch == '"') ? "\">" : "'>"); + (ch == '"') ? "\"" : "'"); if ( buf[buf_index] == '>' ) { /* Unterminated string value? Let's look ahead and test: * (<: unterminated, closing-quote: terminated) */ I don't know why that check is there. Does HTML allow quoted strings to be not terminated? I guess there is a better solution than mine. Greetings Andreas Kemnade

1 0

Something to think about: free vs. proprietary OSes
by Kelson Vibber Sept. 5, 2003

Sept. 5, 2003

<flamebait> So if Dillo is about 100% free software, including the OS it runs on, does that mean we should reject any patches that help Dillo run on Solaris, IRIX, AIX, HP-UX, Mac OS X and so on? If something breaks compatibility with Cygwin, should we let it slide because Cygwin only runs on Windows? What's next, dropping support for SuSE because it's less free than Debian? Red Hat because they call it "Linux" instead of "GNU/Linux?" </flamebait> Where do we draw the line... and why? Compatibility? Ideology? If someone came up with a way to get a working prototype on Mac OS classic, without any effort on your part, would you reject their contribution because the OS is proprietary? If yes, what makes Solaris different? If no, what makes Windows different? A consensus on this issue would be useful, especially the next time someone expresses interest in using Dillo on a nonconventional platform. -- Kelson Vibber www.hyperborea.org

4 3

Re: [Dillo-dev] could not find font
by Jorge Arellano Cid Sept. 4, 2003

Sept. 4, 2003

Hi, > > Hi, I am using dillo for a project here at the univ. In the course of it I > have had to make some modifications to the dillo source so as to be able > to talk with some `tuple space` server. They don't affect the working of > dillo and are abstracted out. Also I need to get the modified dillo on my > ipaq H3670 for use in the project demo. I did the source changes and > everything seems just great. I compiled it for arm using handhelds.org > public clusters and also managed to get it onto my IPAQ. So far so good. > > However the dillo executable doesnt run and bails out saying: > ---- > ** WARNING **: Can't load any ISO8859-1 font!?! :( > ** ERROR **: Could not find any font. aborting... Aborted > ---- This means all the attempts in Dw_style_font_realize() failed. > The dw_style.c seems to look for a font and I would think it fails if it > doesnt find any. That's it. > But what I dont understand is the vanilla gpe > distribution given by the familiar folks also includes dillo and that > works fine on the SAME ipaq. So somehow that dillo executable knows how to > find the fonts? No. It may be that they don't try with the same fontnames! Make sure both provide the exactly the same Dw_style_font_realize() font. If it's the same, check you're using the same dillorc for both. If the problem still persists, using both on the same machine under the same username, using the same dillorc, I'm puzzled! > I am guessing I need to pass a font dir or something > related to x/xfs here? It also works great if I compile it on a desktop > linux box. I am wondering why it cannot find the fonts on the ipaq only. > > Does anyone have any suggestions on how to get this working? Thanks, As above. Cheers Jorge.-

1 0

Dillo on Win2k
by Robin Rowe Sept. 4, 2003

Sept. 4, 2003

FYI, making progress with Dillo on Win2k. http://cinepaint.sourceforge.net/pix/windows/dillo.win2k.gif Cheers, Robin --------------------------------------------------------------------------- Robin.Rowe(a)MovieEditor.com Hollywood, California www.CinePaint.org Free motion picture and still image editing software

9 16

Re: [Dillo-dev] Dillo on Windows fork
by Melvin Hadasht Sept. 4, 2003

Sept. 4, 2003

[Please, let's keep this discussion on the ML] Hi, > > If someone wants to port Dillo to Windows, he > > can look at how Sylpheed-Claws I/O infrastructure was ported to Windows. > > Sylpheed-Claws does not use pthreads, but it forks... > > Windows doesn't fork. That can be emulated with spawn. Threads is a better > approach. I don't know how he did that, but I remember those two parts (IO and forking) were the first things that needed to be ported to Windows. My message wasn't aimed at discussing technical points (I never programmed in Windows), but rather to point to some light-weight gtk 1.2 project that was already successfully ported to Windows, so less experienced developers could take a look on it. > > I think, if someone does not already have the experience of > > porting to Windows.... > > I ported CinePaint to Windows, the largest GTK project ever ported aside > from GIMP. Then disregard my email, it will not help you :). I thought someone was in need of technical help for porting to Windows, so I posted some hints in the case where the poster was not so experienced. > >...looking at Claws versions can be very helpful, as our port > > is done by a very knowledgeable guy. > > Cool, who was that? Just google for Sylpheed-Claws & win32 :) -- Melvin Hadasht

1 0

RE: [Dillo-dev] Dillo on Windows fork
by Eric Gaudet Sept. 3, 2003

Sept. 3, 2003

> Kelson, > > > If Jorge does not want to work on a Windows port, that's up > > to him.... > > That does not necessarily mean the project has to be forked.... > ... > It seems clear that I must take Dillo for Windows elsewhere, > that opinions are too firmly set for Windows users to feel > welcome here soon. It isn't a problem to put my version on > SourceForge. Does anyone here want to get involved now or > shall I go it alone? > That's the beauty fo the GPL: even if the developers of Dillo don't want to help with a feature, a port, you can still do it. And if you end-up adding interesting stuff, Dillo can even get it back! ... > Should I choose a new name for the Dillo fork on SourceForge? > It's not required, AFAIK. The only requirement is to release the sources under the GPL with any release you do. Regarding the need for a windows port, I don't quite see who would need such a thing (besides you, obviously). Windows has phoenix, which is lightweight enough when the OS itself need a big machine only to boot. As for running old versions of windows on a small machine, it really is ludicrous. Running an OS that's unsupported, crash-prone and hopelessly outdated like W9x sounds like a bad idea. This kind of machine is a perfect target for a small linux. Good luck anyway. Eric

1 0

Re: [Dillo-dev] Dillo on Windows fork
by Robin Rowe Sept. 3, 2003

Sept. 3, 2003

Robert, > Well, for me, one of the annoying things about M$ is it's policy of > aggressive obsolescence. Let's not debate the pros and cons of Microsoft. It isn't that interesting and anyone with an opinion is unlikely to change it. In my world, that is, motion picture animation and special effects, the O.S. situation is very different from elsewhere. If you read my articles in Linux Journal then you are aware that in motion picture production Linux dominates as the most popular O.S. on servers and desktops. Windows isn't even a close second. The largest corporate installations of Linux in the world are DreamWorks, ILM, and Disney. To support all operating systems -- even minor ones like Windows -- is a mark of professionalism where I come from. Cheers, Robin --------------------------------------------------------------------------- Robin.Rowe(a)MovieEditor.com Hollywood, California www.CinePaint.org Free motion picture and still image editing software

1 0

Dillo configure cannot find zlib, jpeglib not libpng
by Scorpio Rising Sept. 3, 2003

Sept. 3, 2003

Hello, just joined. I have a configuration problem in Debian stable. The configure file cannot find/use the z, jpeg or png libs that are in /usr/include. *** cannot locate***......disabling jpeg/png/libz I have tried the command line options and even hardcoding the paths in configure.in to no avail. I have installed the dev-headers for these packages. Other programs (GIMP, ShowIMG, etc) find and use them just fine. Dillo works, but no images :-\ Thanks, -- Gregg

2 1

Problems compiling tab+frame patch on Mandrake and RH 7
by Kelson Vibber Sept. 3, 2003

Sept. 3, 2003

I've built RPMs that include version 9 of the tab+frame patch at http://www.hyperborea.org/software/dillo/modified.html (a completely separate page from the mail page with the "vanilla" RPMs). In building these, I ran into problems trying to compile with up-to-date versions of: - Red Hat 7.3 - Red Hat 7.2 - Mandrake 8.2 - Mandrake 9.1 On these systems, after un-tarring the Dillo 0.7.3 source and applying the appropriate version of the patch, make would always fail at the same point: gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I/usr/local/include -I/usr/local/include/libpng12 -I/usr/include/gtk-1.2 -I/usr/include/glib-1.2 -I/usr/lib/glib/include -I/usr/X11R6/include -g -O2 -DXHTML_DTD_FRAMESET -DXHTML_DTD_TRANSITIONAL -D_REENTRANT -D_THREAD_SAFE -Wall -Waggregate-return -c `test -f 'http.c' || echo './'`http.c http.c: In function `Http_query': http.c:167: `VERSION' undeclared (first use in this function) http.c:167: (Each undeclared identifier is reported only once http.c:167: for each function it appears in.) make[3]: *** [http.o] Error 1 make[3]: Leaving directory `/home/kelson/build/dillo-0.7.3/src/IO' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/home/kelson/build/dillo-0.7.3/src' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/home/kelson/build/dillo-0.7.3' make: *** [all] Error 2 The unpatched versions had no problems on any of the above systems. When I checked the patch to see what was changed in http.c, and saw how little there was, I started looking at library versions. At the time I had not tried Red Hat 7.2, so I started up my UML copy. On my first test (just compiling the patched source), it worked. I then checked for and installed updates before trying to build the RPM, and it failed just as the others had. To make a long story short, on Red hat 7.2 I narrowed the problem down to glibc. It worked with glibc 2.2.4-19.3, but not with 2.2.4-32. Unfortunately I can't narrow it down to a version in general. Red Hat 7.3 is on glibc-2.2.5-43, Mandrake 8.2 is on glibc-2.2.4-26.2mdk, and Mandrake 9.1 is on glibc-2.3.1-10mdk. Has anyone else encountered similar problems, and/or come up with a workaround? -- Kelson Vibber www.hyperborea.org

3 3

CPU hog with large selections
by Jorge Arellano Cid Sept. 2, 2003

Sept. 2, 2003

Hi there, Yesterday I came accross this bug while reading: http://freshmeat.net/articles/view/889/ (272KB) If you go to a comment in the middle of the page, start selecting text with the mouse, and move it to the leftmost border (so it selects all the text up to the start), the CPU will hog and dillo will become unresponsive as it was in a dead loop. For instance, from the bottom line it takes 10 minutes to recover on a pentium II 233Mhz. I almost sure this can be improved. Please someone give a hand to fixing it. Cheers Jorge.-

2 2

Dillo used in Puppy Linux
by Barry Kauler Sept. 2, 2003

Sept. 2, 2003

Hi, Introducing myself, just recently subscribed to this list. I'm developing a super tiny Linux distro called Puppy, that runs totally in a 48M or 64M ramdisk, yet has a full suite of GUI apps. The greatest advantage of having everything always in RAM is the fast startup and response. The 48M version 0.6 is on the web, hosted by ibiblio.org, and uses Hacked Links and Dillo. I'm currently using Dillo as the internal HTML viewer. Note, Puppy has a large amount of docs, in HTML. Also using Sylpheed, not Claws. For the "big" Puppy, running in 64M, I've experimented with Mozilla v0.9.8 and Konqueror-embedded. I've retained Dillo as the internal viewer. Note that the small Puppy uses only apps written in plain C, and only has the Athena (neXtaw) and GTK v1.2 libraries -- which is why I want to encourage you Dillo developers to stick with compatibility with the v1.2 library!!! ...and, I know it's dreaming, but if Dillo can get CSS, SSL, and a few other things, it could be the only browser I need. (Flash, SVG, ....) ... small Puppy runs ok on a PC with 128M RAM, but I dream of getting it going on a 64M system. Puppy site: http://www.goosee.com/puppy Regards, Barry Kauler

2 1

table cellspacing and borders
by Ralph Slooten Sept. 2, 2003

Sept. 2, 2003

In my attempts to make a nice design website from pure and propper html, I have discovered that Dillo differs from any other browser when it comes to rendering tables with set widths. Could anyone clarify for me why dillo expands the initially set width of a table when the cellspacing and/or border is increased? What I mean is like this: <table cellpadding="6" cellspacing="10" border="5" width="700" bgcolor="#E3E3E3"> <tr> <td>test</td> <td> test 2</td> </tr> </table> This produces a table which is wider than 700px. If cellspacing="0" and border="0", dillo renders the table fine to it's set width of 700. Is this a bug, or some w3.org specification other browsers like Mozilla / Opera / IE ignore? Using 0.7.3 here if it's any use ;-) Greetings Ralph -- http://axljab.homelinux.org:8080/ "...the software said Win95 or better, so I installed Linux"

1 0

Re: [Dillo-dev] tight loop on IO error
by Jorge Arellano Cid Sept. 1, 2003

Sept. 1, 2003

On Mon, 1 Sep 2003, Andreas Schweitzer wrote: > > All I did was fix the problem for me -- it's possible that there's > > still some object cleanup elsewhere that really needs doing ... > > It almost looks like that there needs some more digging. > > I did some tests comparing Linux and FreeBSD. > 1) The patch needs to be slightly modified for the Linux version, > otherwise dillo won't even start correctly :-) > 2a) When the timeout occurs while contacting unresponsive sites: > The Linux version actually stops eventually and puts > "ERROR: unable to connect to remote host" > in the status bar. However, none of the error debug > messgaes actually occur ... > 2b) Same as 2a but under FreeBSD: dillo still keeps > the "Contacting host" in the status bar, and the > stop-icon is still unfaded. And the error message > gets dumped as indicated in the patch. > > But I'm afraid the logic how all these routines are > accessed is a bit too compliacted for me right now :-) Hi! I just made a quick patch and tested http://primates.ximian.com/~miguel/all.html with a dialup on linux (no CPU hog, no problems, 80 of 83 images got). The idea was to make it analogous to IO_read(). Please test it under BSDs (it's on CVS). Cheers Jorge.-

1 0

tight loop on IO error
by Phil Pennock Sept. 1, 2003

Sept. 1, 2003

When viewing the page [warning, many images]: <URL:http://primates.ximian.com/~miguel/all.html> dillo runs into a tight loop when there's a broken pipe error on retrieving one of the last couple of images. I've tracked down the cause; I'm running the 0.7.3 code, with Frank's patch, but have confirmed the presence of the code in HEAD via the WebCVS interface (wonderful thing). In IO/IO.c:IO_write(), if the write() returns (-1) then unless the error is one of EINTR or EGAIN, the case isn't handled and the "while (1) {}" loop runs wild. With a quick change (including new debug statement) I now see: IO_callback: [GIOcond 4] IO_write IO_write: Broken pipe [errno 32] [St -1] IO_write: aborting: Broken pipe [errno 32] The patch includes some extra cosmetic DEBUG changes which can be safely discarded (and the DEBUG_LEVEL change probably should be ...) -----------------------------< cut here >------------------------------- --- /home/pdp/tmp/zshmiVuTI Mon Sep 1 13:43:24 2003 +++ IO.c Mon Sep 1 13:42:58 2003 @@ -30,7 +30,7 @@ #include "../klist.h" #include "IO.h" -#define DEBUG_LEVEL 5 +#define DEBUG_LEVEL 1 #include "../debug.h" @@ -399,8 +399,12 @@ St = read(io->FD, io->Buf, io->BufSize); io->Status = St; - DEBUG_MSG(3, " IO_read: %s [errno %d] [St %d]\n", - g_strerror(errno), errno, St); + if (errno) { + DEBUG_MSG(3, " IO_read: %s [errno %d] [St %d]\n", + g_strerror(errno), errno, St); + } else { + DEBUG_MSG(3, " IO_read: [St %d] success\n", St); + } if ( St < 0 ) { /* Error */ @@ -446,18 +450,27 @@ St = write(io->FD, io->Buf, io->BufSize); io->Status = St; - DEBUG_MSG(3, " IO_write: %s [errno %d] [St %d]\n", - g_strerror(errno), errno, St); + if (errno) { + DEBUG_MSG(3, " IO_write: %s [errno %d] [St %d]\n", + g_strerror(errno), errno, St); + } else { + DEBUG_MSG(3, " IO_write: success [St %d]\n", St); + } if ( St < 0 ) { /* Error */ io->Status = -errno; - if (errno == EINTR) { + switch (errno) { + case EINTR: continue; - } else if (errno == EAGAIN) { + case EAGAIN: DEBUG_MSG(4, " IO_write: EAGAIN\n"); ret = TRUE; break; + default: + DEBUG_MSG(5, " IO_write: aborting: %s [errno %d]\n", + g_strerror(errno), errno); + return FALSE; } } else if ( St < io->BufSize ){ /* Not all data written */

3 4

Re: [Dillo-dev] Rendering problem
by Jorge Arellano Cid Sept. 1, 2003

Sept. 1, 2003

On Thu, 28 Aug 2003, Ralph Slooten wrote: > On Thu, 28 Aug 2003 12:57:36 -0400 (EDT) > "Andrew A. Gill" <superluser(a)frontiernet.net> wrote: > > > I've been having a problem with dillo rendering some portions of text. > > For instance, the Google sponsored links always take up vertical space > > before the results. > > > > Normally, this isn't something that I really care about, but now it's > > causing slashdot to be rendered as black-on-black text, thus depriving me > > of my ability to read about the new MD5 sums that the RIAA is using. > > Yes, I'm getting this too since today. It's probably a small html error > somewhere, but it's totally skrewed up dillo's rendering to the fullest here. Unfortunately slashdot is sending awful HTML-like content. The black text can be explained because they explicitly set the background, textcolor and visited links to black! (<BODY ...>). Probably it changes with CSS. > > Since about 2 weeks ago I'm also getting a black border on the right (with > exception to today of course where most of the page is black) which just takes > up valuable space ;-) > > Tried the light mode too like Frank suggests, but I prefer the default layout > myself. Anyway, it's not a dillo problem I think, as I get a total of 530 errors > just on the index page itself!!!!! Check now too if ya want: > http://validator.w3.org/check?uri=http%3A%2F%2Fwww.slashdot.org%2F Try with "force_my_colors=YES". Now, it still shows several rendering problems. If those are because of lack of implementation of "floats" we should fix it someday. If it's because of bad HTML, you know our policy. OTOH, I believe slashdot webmasters will not pose major problems to fix their HTML if they're emailed with the facts... Please investigate those and report to them. Cheers Jorge.-

4 5

path for support chinese
by super_hoo Sept. 1, 2003

Sept. 1, 2003

I knowed that it had a patch for dillo0.7.2 to supportting Chinese. Where could I find this patch ? thanks a lot. ______________________________________ =================================================================== µ±µ±ÍøÈ«³¡ÒôÏñ5ÕÛ·â¶¥£¬ÓÅ»ÝŒŽœ«œáÊø£¡ (http://ad4.sina.com.cn/sina/dangdang0825.html)

1 0

Problems with tab patch
by Nigel Wilkinson Aug. 29, 2003

Aug. 29, 2003

Evenin' all I'm trying to patch the CVS with the tab patch but I haven't used patch much. The problem I'm haveing is that I don't know which directory to run the path from, it keeps complaining that it can't find the file to patch Cheers Nigel

2 4

[PATCH]: commands & auto-reload (was Re: dillo / time-controlled reloads)
by Melvin Hadasht Aug. 29, 2003

Aug. 29, 2003

[Henning, you may be interested: the patch implements a --reload option] Hi, To implement the server and remote commands feature in Dillo, I'm going to go step by step. First, I'll implement commands to control Dillo at startup time, then I'll implement the server mode (where a Dillo instance can send the aforementioned commands to another Dillo instance). This is because I may try two differents approaches for the server mode part and because the commands are just a generalization of the command line options and can be implemented separately. Here is my first step: implement commands (like -c "open(URL)"). Patch is attached and is against CVS version. This is a generalization of the command line interface, and in fact, most of the old command line options are translated to a Dillo -c command. The advantage of such approach is that the commands can take optional parameters, like in: dillo -c "open(URL newwindow reload=10 fullwindow=yes)" which is approximately equivalent to: dillo -f -r 10 URL The commands are passed to Dillo as arguments to the -c, --command command line option: dillo -c "open(http://www.google.com)" 1. Command Syntax The argument to the -c command line option is of the form: CMD([STRING][ PARAM=VALUE]...) CMD: command name STRING: first and generic argument: in general a URL or yes|no PARAM: name of optional parameter VALUE: value of optional parameter. Currently, only booleans (yes|no) and integer values are supported. Boolean value can be omitted, in which case the value is considered to be "yes". Integer value cannot be omitted. Parameters are separated with a single space (no commas, to simplify the parser) Commands can be divided into two groups: actions commands, and options commands 2. Action Commands: They do some action: currently implemented commands: open(URL [ fullwindow[=yes|no]] [ reload=SECONDS] [ spamsafe[=yes|no]] [ xid=XID]) parameters speak for themselves. The "reload" parameter is a new feature: it allows to reload the URL each SECONDS seconds. The reload is halted when the page is not viewed and restarted if the user comes back to it. exit() exit the instance. This will be useful when passed to another Dillo instance running as a server. Other commands can be added. The framework is quite flexible and extensible. 3. Options Commands These commands set some options that will affect the following commands. For example, the parameters used in the open() command act only on that particular URL. But it can be useful to be able to set some defaults values. The currently implemented commands that influence these defaults are: fullwindow([yes|no]) reload(SECONDS) spamsafe([yes|no]) xid(XID) all open() commands that come after these options commands will have these default values, unless explicitly overridden. 4. Examples: dillo -c "open(http://www.google.com)" -c "open(file:///home/mhadasht/test.html newwindow spamsafe)" will open two windows: the first with google, and the second with the local file opened in spamsafe mode. dillo -c "reload(10)" -c "open(http://www.freshmeat.net)" -c "open(http://news.google.com newwindow)" will open 2 windows and will reload both URLs each 10s. This is equivalent to: dillo --reload 10 http://www.freshmeat.net http://news.google.com 5. Implementation details: 5.1. Command Line Parsing Command line options other than the help/version options are translated into a -c command and stored in a list. The list of commands is then executed (or, when the server mode will is implemented, sent to another Dillo instance) 5.2. Reload Function a URL can be set to auto-reload itself if it is currently viewed. For that, a new member (reload_interval [milliseconds]) to DilloUrl is added and set with a_Url_set_reload(). Each BrowserWindow can trigger a timer function if the viewed URL (Nav_open_url()) has a non-zero reload_interval. (As soon as another URL is viewed, or when the BrowserWindow is destroyed, the eventually pending timeout function is removed.) The timer is of the one-shot type: when it is triggered, it requests its BrowserWindow to reload the URL, then it stops itself. There is currently no UI to start/stop an auto-reload of a URL, but it is just a matter of using a_Url_set_reload() to a non-zero or a zero reload interval. 6. Remark The -c command and the normal command line options can be seen as two ways of doing the same thing: passing commands to Dillo. But the -c command is more generic and flexible and has a more powerful syntax. 7. Future Plans I plan to implement the server mode. After reading the dpi doc and source, I found that it cannot be used to make Dillo act as a plugin unless we accept to have only one Dillo server. This is because the socket names are tightly linked to the service name. Each service has only one socket. Whereas Dillo servers must have a distinct socket. If the we add another argument to the "start_plugin" cmd: <dpi cmd="start_plugin" msg="dillo" param="SERVER_NAME"> and if we modify the socket naming scheme, then it could be possible. I have to think about that again. But I may still use the communication infrastructure of the dpi. If I find this overkill, I'll revert to the straightforward approach I used before, that is, I won't use the dpi framework. OK, it's late, I hope I managed to be somewhat clear... Cheers. -- Melvin Hadasht

2 5

Rendering
by Jorge Arellano Cid Aug. 29, 2003

Aug. 29, 2003

Hi Sebastian! Hope you're not under a pile of workload... Please review this site: http://cocoon.apache.org/index2.html When loaded slowly (dialup speed) it renders the leftmost column too wide (almost a screen), but Back&Forward fixes it. I you can also give a look at the slashdot problem it'd be great. I'm much more interested in the first though. Regards Jorge.-

1 0

Re: Tab + frame patch, any progress?
by Jorge Arellano Cid Aug. 29, 2003

Aug. 29, 2003

On Wed, 27 Aug 2003, Frank de Lange wrote: > Any progress on the CSS front by the way? Sometime ago we concluded with Sebastian that it may take near three months of work of both of us. We're focusing on other priorities now. I hope we can review your patch soon with him. Cheers Jorge.-

1 0

Running external Gtk programs under Tabs (fwd)
by Jorge Arellano Cid Aug. 29, 2003

Aug. 29, 2003

---------- Forwarded message ---------- Date: Wed, 27 Aug 2003 23:48:14 -0700 From: t <loveall(a)qwest.net> To: Jorge Arellano Cid <jcid(a)softhome.net> Subject: Running external Gtk programs under Tabs Jorge, Please forward this to the dillo/dev mailing list unless you think it is not appropriate. Nice job on dillo, guys! Have been using it since at least 0.2.1. Nice work on the tabs, Frank. They have added a whole new dimension to dillo. Recently, while working on a multi-media project, noticed that I had way too many instances of a text editor open. Couldn't easily find the desired file. About then Frank put up the tab/frame patch. Spent a day or so getting external Gtk apps to run in a tabbed window. This turned dillo into an MDI interface with built-in file picker and HTML help file viewer. This has been working for me in a stable manner, e.g. no seg faults, for about two weeks under heavy use. Which means it will only take someone else about 5 minutes to induce a failure:-). Have modified a file manager, my personal text editor and the erwin HTML editor to run as client Gtk apps, just like dillo. As a matter of fact, dillo can run itself as an external Gtk app in a tabbed window. The patch file for the tab/frame patched version of dillo is set up to do this as a working example. You can download it from: http://www.users.qwest.net/~loveall/xid8-patch.gz !!!! NOTE !!!! This patch is valid only on dillo-0.7.3 patched with Franks version 9 tab/frame patch: http://www.geocities.com/ikbenfrank/dillo-0.7.3-20030821-tabs.patch.gz To modify a typical Gtk app to act as a client requires just the following: In function main(), add these lines at the beginning just after the data declarations: guint32 xid = 0; int i = 1; if (i < argc-1 && !strcmp(argv[i++],"-x")){ xid = strtol(argv[i], NULL, 10); argc -= 2; argv += 2; if(xid == LONG_MIN || xid == LONG_MAX) xid = 0; } This requires that the first two args when calling dillo be '-x <Xid number>'. Find and replace the (typically) single line: main_window = gtk_window_new (GTK_WINDOW_TOPLEVEL); with: if(!xid) main_window = gtk_window_new (GTK_WINDOW_TOPLEVEL); else main_window = gtk_plug_new (xid); See also the working source which references 'xid' in dillo.c and interface.c. Thanks again, to the dillo crew. You guys are awesome. Regards, Terry Loveall

2 1

CinePaint and Windows port
by Robin Rowe Aug. 28, 2003

Aug. 28, 2003

Hi. I'm the project leader for CinePaint, an open source paint program used mainly by the motion picture industry. Our code was branched from GIMP in 1998. How would the Dillo project feel about Dillo being embedded in another application as a help browser? We're looking for a new help system browser and considering Dillo. We are impressed how compact your implementation is. We only need a limited set of HTML features: links, images, tables, bullet lists, number lists, headings, ordinary text, bold, and italic. In these features we need the browser to be absolutely stable. Any issues there? We support all platforms, including Windows. One of the first things I would intend to do is port Dillo to Windows. I saw in your archives that you had made some effort in that direction, but it seemed to peter out. What are the problems? Thanks! Robin --------------------------------------------------------------------------- Robin.Rowe(a)MovieEditor.com Hollywood, California www.CinePaint.org Free motion picture and still image editing software

4 5

Rendering problem
by Andrew A. Gill Aug. 28, 2003

Aug. 28, 2003

I've been having a problem with dillo rendering some portions of text. For instance, the Google sponsored links always take up vertical space before the results. Normally, this isn't something that I really care about, but now it's causing slashdot to be rendered as black-on-black text, thus depriving me of my ability to read about the new MD5 sums that the RIAA is using. (bug 493) -- |Andrew A. Gill |I posted to Silent-Tristero and| |<superluser(a)frontiernet.net> |all I got was this stupid sig! | |alt.tv.simpsons CBG-FAQ author | | | (Report all obscene mail to Le Maitre Pots)| |Nothing here yet: <http://www.needsfoodbadly.com> Temporary sig: -- But thanks for everything else.

3 2

Re: Re: dillo / time-controlled reloads
by Jorge Arellano Cid Aug. 28, 2003

Aug. 28, 2003

On Wed, 20 Aug 2003, Melvin Hadasht wrote: > > Hi Jorge & Melvin, > > Hi Henning & Jorge, > > (Is there a reason why this is not on the ML?) The only reason is that I pressed "reply" on a mail sent to my personal address. :-) > > > > > there's one bit missing. there is NO concept at all about time-based > > > > reloads. the meta refresh header would be reasonably close, but even > > > > better would be something like > > > > > > > > dillo -reload-every 60 http://blah/cgi-bin/blubb.cgi > > > > > > > > it should reload that URL every 60 seconds in that case no matter what. > > > > > > I think reloading it if upon display is also a good choice. I > > > mean, let the user browse other pages without reloading the > > > special URL, and reload it when it's back on display and the > > > timeout has been reached. > > I don't see why a browser should reload an URL if it is not currently viewed, but > I'm not a web expert. > > > > > I looked into adding a sigalrm handler an solving it this way, but > > > > then it turned out that is more complicated > > We can use the glib timeout functions for that. Yes, timeouts are simpler. > > Here is how I would implement the feature: > > - an internal Dillo function to mark/unmark a URL with > "reload periodically when viewed", A simple flag, maybe in the DilloUrl or in the cache entry. > it will also create a timeout > callback that will reload the URL if it is currently viewed. A periodiciy of 0 will > stop reloading. For that, I need to take a look on the browser history/cache, too. Don't worry, I can do that easily (I know pretty well the cache). I'd prefer you to focus on the dpi part. I mean, this dpi will end asking dillo to do the job. For instance: <dpi cmd='set_reload_URL' url='...' always='true'> > > - add a "-r, --reload-every" command line option that calls that function. > In fact, when I finish implementing the "(remote-)command" option, it would be > something like: "-c open(URL, refresh=60)" (that is if I finally decide that > supporting commands with more than one argument is worth the work). Hmmm, I think a simple cli switch to tell dillo to listen to the reloads.dpi is enough: dillo -reloads (or something like it) Then, the client communicates with the dpi socket. All this API is what needs to be designed. I'd appreciate you to work on this part. > > > The first duty is to define what is required for a good, > > > general purpose, kind of reloading dpi. The next one is to reduce > > > the features to a minimum working set, then implement! > > parameters: URL, periodicity (seconds), when (visible|always). This is a good start! For instance: fprintf(reloads_dpi_socket, "push http://localhost/someurl visible\n"); may be all the client program needs to know. Of course it could also be something like: fprintf(reloads_dpi_socket, "<dpi cmd='reloads' url='http://localhost/someurl' visible>" * The first example makes more clear that it goes trough reloads.dpi who parses it and then sends a dpip tag to dillo to request the job done. This may also be handy: "<dpi cmd='reloads' url='http://localhost/someurl' stop>" Note that it may be good to integrate the remote control functionality into the same dpi (and maybe call it remote.dpi?). This means adding a simple and useful API for help programs to use. For instance to let them push new URLs into the same instance of dillo (and not having to start a new dillo per requested page). Ex: The client app. asks the remote control dpi whether there's a running instance of dillo. YES -> send new to-be-pushed URL NO -> start "dillo -rc" (remote control) send new to-be-pushed URL Please polish along these lines. Cheers Jorge.-

2 1

Re: [PATCH]: commands & auto-reload (was Re: dillo / time-controlled reloads)
by Melvin Hadasht Aug. 27, 2003

Aug. 27, 2003

[Please, as this thread is on the ML, do not mail me personnaly (nor CC me). I'm subscribed to the list and will respond when I have time and when I have something to respond.] Hi, > ok, dillo never reloads the images, not even on manual reloads - still > showing days old stats. > you think that is fixeable soon, for some value of "soon"? I don't always have time for Dillo. Maybe this weekend, maybe the next. Concerning the reload time, I should be able to fix it when I have time. Concerning the cached images, I will need more time. Concerning the "void" input driver, is the coredump due to my patch? -- Melvin Hadasht

3 2

tab and frames in webmail
by �޽�� Aug. 27, 2003

Aug. 27, 2003

hi, Frank de Lange: I apply you dillo tab frame patch,and it seem very cool. But when I go to the www.163.com and www.163.net, I found I can't go into my mailbox.It only says that "This page uses frames, but your browser doesn't support them."And the statusbar shows "Contacting host", for ever. In windows IE I can see the soursecode of that page as following: <html><head> <title>网易邮件系统</title> <meta http-equiv="Content-Type" content="text/html; charset=gb2312"> </head> <frameset frameborder=NO rows=80,* border=0 framespacing=0> <frame name="2" src="/newtop.htm" target=contents scrolling=NO marginwidth=0 marginheight=0 frameborder=NO > <frameset border=0 cols=160,* framespacing=0 marginheight=0 marginwidth=0 frameborder=NO> <frame name="folder" src="/cgi/ldapapp?funcid=simplefdr&sid=DBTpkvElTAQAKUwf" target=rtop marginwidth=0 marginheight=0 frameborder=NO scrolling=yes>  <frame name="main" src="/cgi/ldapapp?funcid=foldmain&sid=DBTpkvElTAQAKUwf" target=rbottom marginwidth=0 marginheight=0 scrolling=AUTO frameborder=NO> </frameset> </frameset> <noframes> <body bgcolor="#FFFFFF"><p>This page uses frames, but your browser doesn't support them.</p> </body></noframes> </html> What's wrong? best reguards 　　　　　　　　Jewen_Luo(a)163.com 　　　　　　　　　　2003-08-27

1 0

[lmettler@lamlaw.com: Re: A few remarks on your site layout...]
by Frank de Lange Aug. 26, 2003

Aug. 26, 2003

----- Forwarded message from Lewis Mettler <lmettler(a)lamlaw.com> ----- >Thanks for the corrected index.html page for my site. > >You are correct. Some of those tags did appear to be out of place. > >I first wrote that page 5-6 years ago using a program from Corel. It is >not surprising that it was put up incorrectly. Most likely I was trying >several different formats and then stopped screwing with it. I noticed >it still had the NT label embedded. I have not used NT for many years. > >In any event, I edited the version you provided to replace the NT >reference and updated a few of the tags. > >Thanks again. > >And, good fortune with the Dillo browser. > >Let me know if my site now formats correctly using Dillo. Well, it seems to... If only all sites responded as quickly... Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

1 0

libc dependency in 0.8.0pre (bug 487)
by David McKee Aug. 26, 2003

Aug. 26, 2003

Folks, Updated my CVS at home (Debian 3.0) to 0.8.0pre the other day, and I like what I see. When I tried to do the same at work today (Redhat 6.2), I had problems building. Dpid uses mkdtemp() which was introduced in glibc 2.1.9. However, ./configure has no problems with my glibc 2.1.3. Submitted as bug 487. This constitutes a change from the 0.7 days when (according to the Compatibility web page) any glibc2 would do. Introducing (by hand) a definition of mkdtemp() I found on the web solved all my problems. Browsing the source suggests that you are not in the habit of carrying around lot of little libc bits---and the associated preprocessor code munging---for systems with variant or insufficient libc's. That is fine. It keeps the code base small and simple, but perhaps ./configure should look out for these things. So does this mean some previously working libc's are now going to be excluded, that more conditional compilation stuff will need to go in, or that dpid needs some work to increase it's universal compatibility? -- -- David McKee -- dmckee(a)jlab.org -- (757) 269-7492 (Office)

3 4

On-line demonstration against software patents in Europe
by Frank de Lange Aug. 26, 2003

Aug. 26, 2003

Hi'all, As you might know, the European Parliament will vote on software patents in the EU on the 1st of september 2003. I have added a notice to my patch page (http://www.geocities.com/ikbenfrank) about this. Maybe the main Dillo page can do the same? For more info on this issue, go to http://swpat.ffii.org Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

1 0

Frameset & lanlaw.org
by Frank de Lange Aug. 26, 2003

Aug. 26, 2003

I just noticed that my current version of the tab+frame patch still has problems with sites like lamlaw.com, which put body content outside of frameset content. Even though this is against the spec, I've changed my current version of the patch so it displays the right (frameset) content for sites like this. BTW, lamlaw.com is really weird... The sequence of tags is html - head - frameset - /frameset - noframes - /head - body - /body - /noframes - /frameset - /html. So, it has frameset inside head (wrong), /head inside noframes, an extra /frameset, etc. The document claims to be html-4.0-transitional. Sure. Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

1 0

finger plugin
by Andreas Schweitzer Aug. 22, 2003

Aug. 22, 2003

Hi, as promised in a previous mail. Here is the still quite rough finger dpi intended for interested developers : http://www.hs.uni-hamburg.de/~stcd102/dillo/ http://www.hs.uni-hamburg.de/~stcd102/dillo/adpi-0.1.tar.gz It's configure.in is essentially dillo's configure.in :-) you could actually just put finger.c in the ./dpi/ subdirectory of dillo and adjust Makefile.am accordingly. but now I have an actual distribution :-) In the tarball is also what I call so far a library ... well, or at least the routines that have been re-used over and over. Lastly, there is also an experiment : a term plugin, which lets you call arbitrary programs from within dillo (read the source, though !) Installation: just copy the created .dpi files to ~/.dillo/ and edit ~/.dillo/dpidrc (as long as this file exists), then go to dpi:/finger/ If you want dillo to actually recognize the finger URL you will have to edit src/IO/Url.c and sr/capi.c. Search for the string "ftp" inside these two files and duplicate the relevant part for the finger protocol. Or await how dillo evolves :-) Cheers, Andreas -- **************************** NEW ADDRESS ****************************** Hamburger Sternwarte Universitaet Hamburg Gojenbergsweg 112 Tel. ++49 40 42891 4016 D-21029 Hamburg, Germany Fax. ++49 40 42891 4198

1 0

[PATCH] version 9 of tab+frame patch, frame support 'complete' (?)
by Frank de Lange Aug. 22, 2003

Aug. 22, 2003

Hi'all, The next version (9) of my tab patch is ready (see bottom of this message for download instructions). Frameset support is now almost complete, manual resize has been added, targeted link functionality has been improved, most frame-related attributes actually work, try it I'd say... Frameset (and iframe) support is enabled by default, it can be disabled using the configure option --enable-xhtml-dtd=strict. You can also use --enable-xhtml-dtd=transitional, which will give you iframe but no frameset support. The default option, --enable-xhtml-dtd=frameset, gives both iframe and frameset support. As usual, tab support can be disabled using --disable-tabs. Other new features (since the last version): - CTRL-U pops up the view source window - CTRL-B bookmarks the current page - CTRL-S pops up save page window New preferences frame_blank_in_tab=([NO]/YES): show frames with "_blank" target in new tab (instead of new window) (read the Browser_Frames.txt and Browser_Tabs.txt documents (in the doc/ directory) and the dillorc file for more info on tab- and frame-related preferences (after applying the patch, of course...)) Of course, the patch also fixes some bugs (and possibly introduces new ones). More on frame/iframe support ============================ Frameset support is based on a new widget (GtkFrameset) which implements <frameset> as specified in xhtml-dtd-frameset. This widget is not Dillo-specific, so it can in theory be used in any GTK program. It is a pure GTK widget, not rooted in Dw. It does not provide any special support for Dw-specific features (eg. the full-screen-off-button is not displayed in the widget). I might make a special version for Dillo after the GTK-version has been debugged sufficiently. What works: - loading (nested) framesets, displaying frames at their appropriate location and appropriate size, including margin width/height and scrolling. - loading (nested) iframes, displaying iframes at their appropriate location and with (mostly) appropriate size and alignment. - manual frame resize (row, column or both at the same time (drag border crossing - this does not work between nested framesets) - all frame-related menu options (in the popup menu). Frame menus can be accessed using the 'This Frame...' button in the popup menu, nested framesets have their own 'This Frameset...' menu. This way, Dillo allows you to limit actions to a subset of a framed document, eg. you can search for text in just one frame/iframe/frameset. Top-level menu options usually refer to the whole document (the frameset). An exception to this is the 'Jump To' option, which refers to the frame in which the menu was opened (question: should this contain the anchors for all frames?). Use the 'Show only this Frame' option in the Frame Options menu to get a full-page view of a frame. Same for 'Show only this Frameset'. What does not work yet: - the full_screen_off_button is not shown in frame documents. Use double-click instead to toggle the full screen setting... - the frameborder attribute works, but dw_gtk_scrolled_frame always draws a (focus) shadow around itself - even with the border size is set to 0. This makes it impossible to draw frames seamlessly. Tabs, Frames and browser bloat ============================== version: 0.8.0-pre patch: dillo-20030821-tabs.patch.gz unpatched: 285536 bytes patched, no tabs, no frames: 290444 bytes patched, with tabs, no frames: 298092 bytes patched, no tabs, with frames: 307116 bytes patched, with tabs, with frames: 314892 bytes version: 0.7.3 patch: dillo-0.7.3-20030821-tabs.patch.gz unpatched: 284416 bytes patched, no tabs, no frames: 289708 bytes patched, with tabs, no frames: 297324 bytes patched, no tabs, with frames: 305708 bytes patched, with tabs, with frames: 313452 bytes (gcc 3.2.2, binutils 2.13.90.0.18, CFLAGS "-Os") Judge for yourself if the features are worth their weight in bytes... Download instructions ===================== There are now two versions of the patch, one for Dillo 0.7.3 (release) and one for Dillo 0.8.0-pre (CVS). Make sure to download the correct version for the Dillo source you want to patch! If you patch the CVS source, do not forget to run ./autogen.sh after applying the patch. The CVS patch is made against 20030820 CVS, while the 0.7.3 patch is made against Dillo-0.7.3.tar.bz2 (the source tarball on the main Dillo website). Both versions of the patch can be downloaded from the usual (ad-happy, but...) Geocities site: http://www.geocities.com/ikbenfrank/ More specifically for both versions: Dillo 0.8.0-pre (CVS) ===================== http://www.geocities.com/ikbenfrank/dillo-20030821-tabs.patch.gz size: 66661 bytes (gzipped, as downloaded) 281416 bytes (uncompressed) md5sum: 6afb73d97d3db88c629726cef100f604 (gzipped, as downloaded) c3b6ee3095d4b230a78cf836516e0023 (uncompressed) Dillo 0.7.3 (release) ===================== http://www.geocities.com/ikbenfrank/dillo-0.7.3-20030821-tabs.patch.gz size: 93929 bytes (gzipped, as downloaded) 399156 bytes (uncompressed) md5sum: 0363c9ad9f94df9b1bca23e5489e5c5e (gzipped, as downloaded) 8fca8c49a01cd6a6103661720eb8d60d (uncompressed) As always, check the site for the most recent version: That's all for now, Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

2 1

Some points about the dpi's
by Andreas Schweitzer Aug. 22, 2003

Aug. 22, 2003

Hi, All right, I did it. I made my first dpi ! :-) (heavily based on the hello dpi, though). I will poilsh it a bit and release it soon. It can handle finger URLs. Two caveats : there is no standard for finger URL's, only some drafts http://www.watersprings.org/pub/id/draft-ietf-uri-url-finger-00.txt http://www.watersprings.org/pub/id/draft-ietf-uri-url-finger-01.txt http://www.watersprings.org/pub/id/draft-ietf-uri-url-finger-02.txt http://ftp.ics.uci.edu/pub/ietf/uri/draft-ietf-uri-url-finger-03.txt I implemented version 01 and 03 (if you check the documents you'll know why). However, I did find real finger URL's in the wild. The second caveat : dillo does not automatically call this plugin upon such a URL. It is easy to patch dillo, though. However, this brings me to my questions : 1) will there be a generic URL->dpi mapping ? I could imagine that dpid could actually handle that. I think, one would need to only modify IO/Url.c and capi.c to simply pass through EVERYTHING to the dpi framework. However, some "core URL handling dpi's" that will become part of the dillo package can probably stay hardcoded in dillo. 1b) Either way, I am planing on writing a "catch-all-URL handler" or "fallback handler" or "univeral URL handler" or the like. It will give the user the possibility to start an external program for handling certain URL's. In fact, with my current dpi, this is quite easy to do :-). 2) Does it make sense to write a library for all the dpi's ? I, for myself, did start simply collecting all the functions that I found in dpi/ and are used by many dpi's and put them in a library. Like the various send_stream functions or the get_attr_value function, or the encoding and decoding functions. 3) About configuration (and I think I read comments about this in the code), shouldn't there be only one config file ? and not dpidrc and possibly different configs for all the dpis ? maybe even merge cookiesrc with dillorc ? I guess some of this (or all of it ?) is being worked on by some of you. Maybe this mail helps to avoid duplication :-) Cheers, Andreas P.S.: I also have some ideas about improving the ftp plugin. There is a pretty simple ftp library, and I think it does make sense to actually log in an ftp server and stay logged in for all the operations. (unlike wget which logs in and out for every file). This is not only faster but helps with busy ftp sites, too ! -- **************************** NEW ADDRESS ****************************** Hamburger Sternwarte Universitaet Hamburg Gojenbergsweg 112 Tel. ++49 40 42891 4016 D-21029 Hamburg, Germany Fax. ++49 40 42891 4198

3 3

[PATCH] Tentative solution of #448, #449
by tomas＠fabula.de Aug. 20, 2003

Aug. 20, 2003

Hi, I've got a problem with 8 bit depth visuals (the images have the wrong palette). This is similar to bug #449 of your bug tracking database (thus I tentatively offered to fix it. Tentative patch is below. Besides, this looks very much like bug #448, so this might be related. I call the patch ``tentative'' because (a) I don't really know 100% what I'm doing here (b) I haven't tested this on other color depths yet. Nevertheless images show correctly now; the price I pay for this is Technicolor. Regards -- tomas ------------------------------------------------- diff -Naur dillo-0.7.3/src/dillo.c dillo-0.7.3-color/src/dillo.c --- dillo-0.7.3/src/dillo.c Tue Jul 22 22:51:22 2003 +++ dillo-0.7.3-color/src/dillo.c Fri Aug 8 19:01:32 2003 @@ -245,6 +245,9 @@ signal(SIGCHLD, Dillo_sigchld_handler); + gtk_widget_push_visual(gdk_rgb_get_visual()); + gtk_widget_push_colormap(gdk_rgb_get_cmap()); + /* check that ~/.dillo exists, create it if it doesn't */ dir = a_Misc_prepend_user_home(".dillo"); Dillo_check_home_dir(dir); @@ -304,6 +307,9 @@ a_Prefs_freeall(); a_Dw_freeall(); a_History_free(); + + gtk_widget_pop_colormap(); + gtk_widget_pop_visual(); g_print("Dillo: normal exit!\n"); return 0;

2 1

Re: Another review of Dillo 0.7.2
by Jorge Arellano Cid Aug. 18, 2003

Aug. 18, 2003

On Sun, 17 Aug 2003, Pete Spicer wrote: > Hi there! > > There was a review of Dillo 0.7.2 in Linux Format issue 42 (July 2003) which > featured it in their "Hot Picks" section (it got a half-page review with > screenshot browsing Slashdot.org on page 41) Good! Thanks for the information. > It was written by Mike Saunders, and quotes it as "starting up amazingly > quickly" and "works marvellously well for simple browsing tasks and the > rendering engine rivals Links for pure speed". It'd be very nice to have a link to it, but I don't know whether LinuxFormat has it online. Does anyone know the link? > Oh, and by the way, I use it myself on my machine when administering my > database with phpMyAdmin (just because it uses frames doesn't mean it can't > cope... provided you pick the right frame to be displayed, Dillo is fine > with PMA!) This is also very good to know. We may surprise you again soon! > Thanks for a great browser, > Pete :-) Cheers Jorge.-

1 0

dillo - my favorite browser (fwd)
by Jorge Arellano Cid Aug. 18, 2003

Aug. 18, 2003

---------- Forwarded message ---------- Date: Thu, 14 Aug 2003 20:57:41 +0200 From: Mario Moder <clay-man(a)freenet.de> To: jcid(a)softhome.net Subject: dillo - my favorite browser Hi! Congratulations! I have been using Mozilla Firebird 0.6.1 and it is very disappointing that it takes half a minute to start on my good old P1-200 with 64 MB RAM. Even when it is running, it runs really slow and uses a lot of memory. I tried the new dillo 0.7.3 and, wow, it started in one second and it renders 90% of the pages I visit everyday quite good and fast. I really look forward to watch this browser grow more mature but stay small and fast. Best wishes, Mario

1 0

Dillo hang as use chinese font
by Siu Kam Ming Aug. 18, 2003

Aug. 18, 2003

Hi , I was using the attached patch to show chinese character in dillo. It worked fine. I can use a very fast browser to browse my favorite chinese webpage. Since 0.7.2, dillo become unstable after applying this patch. As I selected any text in the rendered area, dillo hang. Does someone know the reason or the solution? Best Regards, Ming

2 1

Version 8 of tab+frame patch: smallproblemwithwhitespaceinmultilength
by Frank de Lange Aug. 13, 2003

Aug. 13, 2003

Hi'all, I just noticed that the current version of my tab+frames patch does not handle multilengths with whitespace correctly. I already fixed it in my version, will include the fix in the next version of the patch. If you encounter framesets which look weird, it may be caused by this problem. As the multilength parser was based on the one found in html.c, this problem is also present in anything which uses Html_parse_multi_length_list(), or which feeds length strings with leading whitespace to Html_parse_length_or_multi_length(). So, a multilength like "*,*" works fine, as does "* ,*", but "*. *" does not work correctly. As far as I can see, these functions are used in tables and embedded widgets. My fix in gtkframeset.c was to rewrite the parser into something slightly more robust (using g_strsplit and g_strstrip). Something similar is needed for html.c. I might include this in my patch, but unpatched Dillo also needs a fix so it is better solved in CVS. Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

1 0

RPM now available for SuSE 8.1!
by Kelson Vibber Aug. 13, 2003

Aug. 13, 2003

With the assistance of Patrick Shanahan, a Dillo 0.7.3 RPM suitable for SuSE 8.1 is now available! http://www.hyperborea.org/software/dillo/ Kelson Vibber www.hyperborea.org

1 0

[PATCH] Version 8 of tab+frame patch is out
by Frank de Lange Aug. 13, 2003

Aug. 13, 2003

Hi'all, The next version (8) of my tab patch is ready (see bottom of this message for download instructions). Frameset support has been upgraded to an 'almost usable' level. The major missing feature is manual frame resize support, which will come in a later version. Frameset (and iframe) support is enabled by default, it can be disabled using the configure option --enable-xhtml-dtd=strict. You can also use --enable-xhtml-dtd=transitional, which will give you iframe but no frameset support. The default option, --enable-xhtml-dtd=frameset, gives both iframe and frameset support. As usual, tab support can be disabled using --disable-tabs. Other new features (since the last version): - CTRL-R will reload a page (or frameset) Of course, the patch also fixes some bugs (and possibly introduces new ones). Please, test this patch and tell me what breaks. Or what works. Or whatever. More on frame/iframe support ============================ Frameset support is based on a new widget (GtkFrameset) which implements <frameset> as specified in xhtml-dtd-frameset. This widget is not Dillo-specific, so it can in theory be used in any GTK program. It is a pure GTK widget, not rooted in Dw. It does not provide any special support for Dw-specific features (eg. the full-screen-off-button is not displayed in the widget). I might make a special version for Dillo, which does implement these features. What works: - loading (nested) framesets, displaying frames at their appropriate location and appropriate size, including margin width/height. - loading (nested) iframes, displaying iframes at their appropriate location and with (mostly) appropriate size and alignment. - all frame-related menu options (in the popup menu). Frame menus can be accessed using the 'This Frame...' button in the popup menu, nested framesets have their own 'This Frameset...' menu. This way, Dillo allows you to limit actions to a subset of a framed document, eg. you can search for text in just one frame/iframe/frameset. Top-level menu options usually refer to the whole document (the frameset). An exception to this is the 'Jump To' option, which refers to the frame in which the menu was opened (question: should this contain the anchors for all frames?). Use the 'Show only this Frame' option in the Frame Options menu to get a full-page view of a frame. Same for 'Show only this Frameset'. What does not work yet: - the full_screen_off_button is not shown in frame documents. Use double-click instead to toggle the full screen setting... - manual frame resize is not supported yet. - the noresize and scrolling attributes are not supported yet Tabs, Frames and browser bloat ============================== 0.7.3 unpatched: 285152 bytes Patched, no tabs, no frames: 289356 bytes Patched, no tabs, with frames: 303052 bytes Patched, with tabs, no frames: 297100 bytes Patched, with tabs and frames: 310828 bytes The frameset implementation has grown about 10 Kb since the last patch. This is mainly due to the implementation of frame size support and the shift to a new widget (which has to be included in the binary, as it is not part of GTK (yet?)). Anyway, tabs alone cost about 12 Kb, frameset alone approximately 18 Kb, both together about 25 Kb. Bear in mind that the code has not been optimized for size yet. Download instructions ===================== There are now two versions of the patch, one for Dillo 0.7.3 (release) and one for Dillo 0.8.0-pre (CVS). Make sure to download the correct version for the Dillo source you want to patch! If you patch the CVS source, do not forget to run ./autogen.sh after applying the patch. The CVS patch is made against 20030813 CVS, while the 0.7.3 patch is made against Dillo-0.7.3.tar.bz2 (the source tarball on the main Dillo website). Both versions of the patch can be downloaded from the usual (ad-happy, but...) Geocities site: http://www.geocities.com/ikbenfrank/ More specifically, the CVS version can be found here: http://www.geocities.com/ikbenfrank/dillo-20030813-tabs.patch.gz size: 64875 bytes (gzipped, as downloaded) 272722 bytes (uncompressed) md5sum: a1315001bff22326b33a96b3828341b3 (gzipped, as downloaded) eed9681876272e81f161e254b684fd85 (uncompressed) The 0.7.3 version is here: http://www.geocities.com/ikbenfrank/dillo-0.7.3-tabs.patch.gz size: 117970 bytes (gzipped, as downloaded) 544402 bytes (uncompressed) md5sum: 93b55564ad4b18367f9632c4616e5890 (gzipped, as downloaded) c84a87687950d0e6670e01ea377d3ee7 (uncompressed) As always, check the site for the most recent version: That's all for now, Cheers//Frank -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ] -- WWWWW ________________________ ## o o\ / Frank de Lange \ }# \| / +46-734352015 \ \ `--| _/ <Hacker for Hire> \ `---' \ +31-640037120 / \ frank(a)unternet.org / `------------------------' [ "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est." ]

5 6

[patch] Remote Control
by Melvin Hadasht Aug. 13, 2003

Aug. 13, 2003

Hi, I just made a patch to add remote control for Dillo. 1. Summary: 1.1 Why? To be able to command Dillo from another process. E.g.: if Dillo is used as a help doc browser of a program, the program can implement context sensitive help and associate to each widget a help URL. The same Dillo instance can then be used to show the differents help docs. Other example, which also was the motivation of this patch: Sylpheed-Claws uses Dillo to render HTML mail, but currently it has to run a new instance each time the user views an HTML mail. Having only one instance is more efficient and improve responsiveness. 1.2. HowTo To be remote controlled, a Dillo instance must be explicitly started with a special command line option (-s, --server NAME), e.g.: dillo -s server A client instance sends commands to this server by referencing it and by using the -c, --command options, e.g.: a. dillo -s server -c "open(http://www.google.com)" -c "new-window(www.av.com)" b. dillo -s server http://www.google.com amounts to: dillo -s server -c "new-window(http://www.google.com)" Hence, running dillo -s server -c "open(URL)" mutliple times will launch only one instance of dillo at the first time, and the subsequent calls will just hand the arguments to the server. This is more practical than Mozilla's current behaviour where the user has to use two different commands to start and to control a server. The list of currently known commands is provided by the -H, --help-commands option: dillo -H Commands that can be used with the -c, --command option: open(URL) Open URL in current window. new-window(URL) Open URL in a new window. Exit() Terminate browsing session and exit. Command names are case insensitive. It is allowed to have more than one command option. New commands can be added, the syntax being: command(.*) 2. Ressources In dillorc: server_socket_path by default, it is "/tmp". It is used to create the socket name which is: $server_socket_path/dillo-$USER-$NAME. where $NAME is the server name from the -s, --server NAME option. E.g.: /tmp/dillo-john-help 3. Caveat If gtk specific command lines arguments are given, arguments are messed around, due to the fact that the patch delays gtk_init() after private arguments parsing, This will be fixed in the next version If a Dillo server does not quit cleanly, the socket corresponding to the server is not removed. Subsequent attempts to run a server with the same name will fail until the socket is removed (by default: /tmp/dillo-$USER-$SERVER_NAME). A solution is currently searched. The -c "exit()" works only when used from the client side The -c commands are currently ignored if no -s option is given. Any comments are welcome. Especially use cases for the remote control. 4. Details 4.1. Becoming a server Communication between server and client is done via UNIX domain socket. The advantage between this and FIFO being that a separate connection is made for each client, so communcation are guaranteed to not mixup. FIFO, on the other hand, can mixup communication of multiple clients if the messages are larger than some value (PIPE_BUF). This patch uses dynamic allocation for the messages received by the server. When the -s, --server NAME option is given, dillo checks wether a socket correspondig to that name is available and responding. If yes, it will hand any -c, --command CMD and any URL|FILE it finds in its command line options to that running server. URL|FILE are interpreted as "-c new-window(URL|FILE)". If no server socket is responding, it turns it self into a server by creating that socket. After that, it handles the -c, --command CMD and the URL|FILE options itself. 4.2. GUI initialisation The gtk_init() function is now called after dillo options parsing to speed up client execution. Indeed, a client instance does not need to initialise the GUI. This introduces the problem of messed up args if gtk args are given. This could affect the URL|FILE arguments. A solution would be to save them in a separate list for later processing. Another solution would be to interprete the URL|FILE arguments as -c open(URL|FILE) at an early stage. This will also help fixing the problem of the -c commands being ignored if no -s option is given. 4.3. Commands Commands are simple of the form: command(arg..). Arguments are parsed by the shell, so the user must take care that the command is interpreted as a single argument by the shell. Arguments should not contain unescaped parenthesis or spaces. Therefore, the following example is incorrect: dillo -s server -c open(file:///home/user/My documents) because: 1. the shell interprets the parenthesis, and 2. there is a space in the argument. This should be escaped in the following manner by using quotes: dillo -s server -c "open(file:///home/user/My documents)" It is easy to add new commands. -- Melvin Hadasht

3 4

Recovering health.
by Jorge Arellano Cid Aug. 12, 2003

Aug. 12, 2003

Hi there, Sorry about being absent from the list the past few days. I caught what here's called "rotavirus", it starts with stomach problems and when you recover, it knocks you with a heavy cold! That's several days, staying in bed, etc... (I'm now recovering from the cold). What we were doing? Ah yes, trying to make the 0.7.3 release. Well, as the bug discovered requires an extensive revision of the incremental resize mechanism, we decided to tag current CVS as 0.7.3-pre and upload 0.8.0-pre there and to continue as if it was released. Once there's a patch that fixes the bug, 0.7.3 will be released to the public, but in the meanwhile we'll be working on 0.8.0-pre. I updated the dpi1 spec (to be commited after this email), and hope to upload my 0.8.0-pre tarball before Sunday. Thorben: I sent a couple of emails to your auriga address, but got no answer. Maybe they were filtered as SPAM (some servers filter softhome, and I don't blame them!). That's why I'm writing to you here. Did you receive them? How is the email accounts enabling process at auriga going on? Ah, and I received a couple of SPAM rejection notes from external servers for SPAM delivered by our list. Is it possible to hook SpamAssassin to dillo-dev so it filters out SPAM? Best Jorge.-

2 1

Installation Problem
by William Trenker Aug. 12, 2003

Aug. 12, 2003

I'm getting the following error when installing Dillo. /packages# rpm -Uvh dillo-0.7.3-1.rh7.i386.rpm error: failed dependencies: libpng.so.2 is needed by dillo-0.7.3-1 /packages# l /usr/lib/libpng.so.2 lrwxrwxrwx 1 root root 17 May 5 21:05 /usr/lib/libpng.so.2 -> libpng.so.3.1.2.1* I'm not sure what my next step should be. I'm wondering why libpng.so.2 is linked to a file called libpng.so.3? Do you think I should try and find an RPM for the libpng.so.2 library and reinstall that? Where do you think I would find such an RPM? I have had problems with web pages that have PNG's in them ever since Dillo 7.2 but everything was fine with Dillo 7.1 so I'm not sure what I should do. Thank you for any help you can offer. Bill -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/

3 2