Saturday, August 13, 2011

TenFourFox 6: saved by the sneakernet

UPDATE: retracted due to issue 82, respin available tonight or tomorrow

I had a brainwave (well, more like a duh moment) over the weekend about how to sneakernet the Firefox 6 repo updates to the G5 build system, which is still marooned from the Internet. (A parenthetical note about that: I've sent Time Warner the beginning pieces of a lawsuit and basically told them I'm walking away from the contract unless by some miracle the line's in this week. I'm quite certain it won't be. I'm getting bids for a T1, which I probably should have done in the first place; with luck, the downtime won't be more than a week or two more, and then Floodgap will be back in business with a super duper faster line.) So the iBook made several trips to the local Starbucks and back to the house, and several hateful cups of Zen tea later, TenFourFox 6.0 RC/final is built and this post is being typed in it.

I'm a little nervous about this release; there's been pretty much no time for serious conformance testing on all the architecture builds, although the iBook G4 checks out and so does the G5 (what I can test locally). So please thrash it. There's no new features in it, just fixes. The changesets for beta 1 still apply, so there is no new changeset package. If there is a serious showstopper, I did do a cursory security review of Firefox 5, so we can build a 5.0.2 for security parity purposes if needed.

For users new to TenFourFox 6, remember: plugins are disabled (see previous posts).

Looking ahead, Tobias (bless him) has already done just about all the porting work on Fx7 in advance and he says it seems to run really well on his G4. I have his changesets on one of the marooned servers, so as soon as I can snatch the Fx7 repo once it goes beta, we will work on that. I am quite concerned about Fx8, though. Mozilla has landed several changes to the build system that expect builds to occur on 10.6 (even though the build is ostensibly linked against the 10.5 SDK) and obliterate PPC references. This is unwindable (though it will probably hit the vanilla PPC builders such as MonkeyBoy and El Furbe), but may have other side effects, and a simultaneous change to require and use CoreUI for drawing UI elements is also of serious alarm to me as it means our Tiger widget code may no longer work even if we restore the old code. They seem to be having various issues with it, however, as you might expect from an Apple private interface, and I'm hoping they give up and punt it to Fx9.

Watch for a re-release tonight or tomorrow Pacific. Read the release notes, then grab:
  • G3
  • G4/7400
  • G4/7450
  • G5

26 comments:

  1. Quick Leopard Alert:

    There is a graphics glitch in 6.0 that causes the progress bar (maybe other UI elements) to be drawn 1 pixel off in the active (foreground) window.

    http://postimage.org/image/1r7jc7zs4/

    When the window is in the background, it's drawn correctly. When the window is switched from active to background and back to active, the glitch is corrected. Anything that draws from then on is 1px off again (see screenshot). As always, don't have an Intel Mac to verify, but I can open a bug. Happens on G4 7450/10.5.8. Doesn't happen on G3/10.4.11.

    ReplyDelete
  2. So it seems there will be quite some work to be done to backport Fx8.

    Fx7.0a2 has been used as the only browser on a PB G4 here since I did the first release build of it. Apart from distortions in WebM videos (which only occur in release but not in debug builds) there hasn't been any problem so far. One time it didn't respond to mouse clicks any more with everything else still working prooerly.

    ReplyDelete
  3. Chris, go ahead and file it with screenshots of all the elements affected and I'll look at it. It's not a showstopper per se but we should get a fix in for it if we have to respin for some other reason. Did it happen in beta 1?

    Tobias, yeah, I'm worried about Fx8, but then we didn't think we'd make it to Fx5 and here we are at 6. Still, if CoreUI becomes a hard dependency we may have a lot of work to do for future porting efforts, and more so when Mozilla drops 10.5.

    ReplyDelete
  4. It looks like only the progress bar is affected whereever it occurs (downloads manager, the weird black "window" that tells you the download progress for add-ons, check for updates etc). It happened in 6.0RC/final. I've been using 6.0beta1since it was out and have never seen it there.

    ReplyDelete
  5. So far on my eMac using the 7450 bulids everything has been ridiculously snappy.
    Running things plug in free and the rest of the back end tweaks seem to be just what the doctor ordered performance wise.

    Safari 5 is around should somebody need something in Flash but other than that that's about it.

    ReplyDelete
  6. Wonderful ! My G7450 will be sooooo happy ! The only show-stopper for me are the hateful cups of Zen at Starbucks. Must now rejig my internal systems for a 'Latte' build :P

    BTW : how can i be a member of tenfourfox blogspot ? Can't seem to find the key. Ta. jim

    ReplyDelete
  7. First, let me say that I appreciate all of the effort being put into this project. Has anyone experienced printing problems with 6.0 RC? The RC broke printing to an HP printer and to PDF files for me. Switching back to 6.0b1 restores printing functionality. Printing to a PDF produces a file, but Preview reports "File error. Couldn't open the file." The PDF produced by the RC is non-zero in size, but smaller than the working PDF produced by 6.0b1.

    ReplyDelete
  8. rardin, is it only with HP printers or is it with any printer? The only thing that would affect printing that landed after b1 is bug 665218. There is a test case there for people who want to see if they a) crash b) can print. If you crash, but can print, in b1, then maybe we need a different fix for that (or back out the fix entirely?). However, I'd like to rule out a driver issue, though the .pdf part implies it is the application.

    I'm absolutely befuddled on the progress bar issue, though: the shortlog http://hg.mozilla.org/releases/mozilla-beta/shortlog doesn't show anything obvious that would have affected it. If someone is able to find a bug number, that would really help -- I'm still looking.

    ReplyDelete
  9. (oops, posted too soon)

    turtle, glad to hear the performance is good :)

    jim, I'm thinking next time I'll simply drink Chai. The baristas at this Starbucks have no idea what Earl Grey is. As far as membership, that's controlled by Google -- I don't know how it works.

    I'm busy running hg histories on the widget library but I can't find any recent changes to prog bars still. Mystifying.

    ReplyDelete
  10. We're an HP shop when it comes to printers. I can say that it's failed to print in the same way on 3 different LaserJet models (4350, P3015 and a Color 4600). The job seems to be sent to the printer normally (the printer icon appears in the dock, then disappears and, in the case of the 4600, the printer received something, because it awakened and calibrated itself), but nothing is ever produced on the printer. As I mentioned, printing to PDF produces a smaller file than 6.0b1 that cannot be opened by Preview or Acrobat. I noticed these problems on a G4-based Mac running 10.4.11. I just tried a test print to the 4350 using 6.0RC on a Power Mac G5 running 10.5.8 and it worked fine, so this may be a Tiger-only issue.

    ReplyDelete
  11. rardin, can you see what happens in bug 665218 with the test case on the 10.4.11 and 10.5.8 machine? Note that it may crash the browser; this is intentional. Let me know if it crashes/prints in each combination.

    Regrettably I'm on my office Windows PC, so I can't test this myself right this moment.

    ReplyDelete
  12. Assuming that I'm supposed to open the test case for bug 665218, then try to print the blank page that results, it doesn't crash either b1 or RC on my Mac running 10.4.11. A blank page is ejected from the printer when I test using b1. The test using RC behaves similarly to what I described previously: the job seems to be sent to the printer, but no blank page is ejected from the printer.

    ReplyDelete
  13. I have independent confirmation of the bug from another user, but rardin, if you can test that test case for me I can have this backed out today and spin out new builds to upload tonight or tomorrow. This officially counts as a showstopper, so the RC is retracted.

    If anyone finds a guilty party for the progress bar issue I'll include it too, but I'd prefer not to hold the release for it past tomorrow.

    ReplyDelete
  14. >Has anyone experienced printing problems with 6.0 RC?

    I have no problems printing to PDF or to my trusty Kyocera FS-1010 with 6.0RC/final (PowerBook G4, 10.5.8).

    ReplyDelete
  15. Ah, I just see you did. That confirms that's the bug. Thanks!

    ReplyDelete
  16. Chris, could you print fine in 6.0 b1 also? If so, probably the safest fix is to back the bug out entirely (and just let it crash in those edge cases). For 7.0 we can work on a conditional fix that does one thing in 10.5 and the original behaviour in 10.4.

    ReplyDelete
  17. No crashes on the G5 running 10.5.8, either. A blank page was ejected from the printer when I used either b1 or RC.

    ReplyDelete
  18. This comment has been removed by the author.

    ReplyDelete
  19. We'll just back that bug out then. Issue 82. Thanks, rardin and Chris!

    ReplyDelete
  20. I could print and save to PDF just fine in 6.0b1. I do this a lot, so I'm sure I would have noticed it.

    ReplyDelete
  21. is there any way a user can enable Flash Player in TenFourFox 6? i have ben to 30+ webpages on safari that still work with flash 10.1. but i wood like to use TenFourFox for its speed and fullscreen mode but wood like flash is there any way?

    ReplyDelete
  22. Set tenfourfox.plugins.enabled to true in about:config.

    ReplyDelete
  23. "Set tenfourfox.plugins.enabled to true in about:config."


    Thank You! thank you! thank you!!!!!!!!!!!!!!!!!!!!!!!!! now tenfourfox 6 is my only web browser and i am loveing it!!!

    ReplyDelete
  24. I too found my way to the tenfourfox.plugins.enabled setting, and this comment thread confirms it works. Question tho: HOW? I've changed the setting and it appears to be holding, but still getting the "plugins disabled by default" message and no Flash. Are there other settings that need to be flipped? Or a restart required? Ah'mina try the latter...

    ReplyDelete
  25. I am a little bit disappointed. I did realize that plugins were going to be disabled in version 6, but I had no clue Flash would be annihilated too. I avidly watch YouTube videos, and most of the content worth watching has adds in it which means that I can't view it through HTML 5. I hope that this can be resolved in the next version. If not, I guess I'll say bye to TenFourFox. I rarely ever used my PowerMac anymore, so it's not too hard of a goodbye.

    ReplyDelete
  26. Would greatly appreciate if you could do the 5.0.2 patches thing, Cameron. I will probably eventually give over to 6.0, unless something else turns up but for now need more transition time to get used to the idea of no flash...

    ReplyDelete

Due to an increased frequency of spam, comments are now subject to moderation.