ImproveMathEquationEditor/IRCMeetings/June2010/21st July2010

This page is the IRC log of the 21st July 2010 IRC meeting ( #education.openoffice.org ) with students for Google SoC 2010.

Back to July 2010

Attendees :


 * Fridrich Strba (Novell)
 * Mathias Bauer (Sun-Oracle)
 * Eric Bachard (OOo Education Project
 * Michal Spisiak (student, Google SoC)
 * Jonas Jopsen (Student, Google SoC)

-

[10:59] * tl-away is now known as tl13

[11:01]  mba: tl13: jopsen: hi, do we want to start?

[11:01] fine with me

[11:02] Spiso: okay with me... Have anybody got an agenda ?

[11:02] +1

[11:02]  not really

[11:02]  I just know that ericb2 wanted to integrate my work into openoffice

[11:03]  and therefore I wanted to ask for help with writing specs, or whatever is needed to do it

[11:04] * Fridrich (~fridrich@166-249.198-178.cust.bluewin.ch) has joined #education.openoffice.org

[11:04]  News from cws: txtl10n: created

[11:04] Is the coding already done? Or do you want a spec to see what still might be required to do?

[11:04]  tl13: mba: the code I wrote basically retrieves the baseline when formula is created and then aligns the object 'From Bottom' using this value so that baseline in formula is identical to the one of text around

[11:04] Hi

[11:05] I'll have to quit for a while

[11:05] but I can propose an agenda

[11:05] 1) point on the baseline

[11:05] 2) point on the other tasks

[11:05] 3) discuss about prepare a plan for specs

[11:06] Spiso: perhaps we should start with having the code in a CWS so that it actually compiles and runs in an upstream environment. Then we can give feedback on it and discuss possible UI changes.

[11:07] see you in ~15min

[11:08]  mba: there should be a branch created for starmath changes in some time, I'm not sure whether this is already in progress, Fridrich might be able to tell us more

[11:08]  mba: what UI changes would you expect?

[11:09]  yeah, I will do a CWS when the GSoC is succesfully done

[11:09] I expect UI changes as the user somehow needs to activate the new alignment.

[11:09]  Fridrich: but as mba said, it might be worth doing now, so that he can give better feedback, and I can do some changes _before_ the coding is done

[11:09]  and "discussing changes" can kill all development, from personal experience

[11:10] Fridrich: I don't expect this to happen. ;-)

[11:11] I want to get a common understanding about the new functionality. Changing the UI should be rather easy - but it should be agreed upon.

[11:11]  mba: Do you imagine that for new documents user has choice to use the old and the new alignment?

[11:12]  another load of compatibility options? God protect us!

[11:12]  there certainly needs to be something for old documents that would calculate the baseline for all equations and aligns them if required

[11:13] UI changes maybe something like :

[11:13] a) maybe right click context menu on formulas, and then a new entry in the 'Alignment' sub menu

[11:13] b) definitely a new entry in: right click context menu, 'Object' then 'Type' tab page in the list box for 'Vertical' alignments if the anchor is set to 'As character'.

[11:13] At least until now I was expecting to give that choice, yes. Otherwise we had to discuss which alignment should get replaced and how old documents shall be treated then.

[11:14]  makes sense

[11:14] The new compatibility option would be needed in case we do *not* provide new UI and instead of that automatically reinterpret an existing alignment.

[11:15] If we think somewhat more large scale I'd like to see a "Tools/Math" entry with a sub menu. There could be an entry to align all formulas in the document to base line. In the long run

[11:16]  there is no problem with old documents, if you open an old document everything stays the same, the algorithm that aligns it has to be activated, for which there could be an UI thing

[11:16] <Spiso> for instance as tl13 said

[11:16] Spiso: yes, that's what I was talking about.

[11:16] In the longer run, I discussed with FL that something like "change base font height", "font type", ... for math objects would be useful as well. But thtas the long view only...

[11:17] Spiso: do you have a preferred way to activate the new alignment? Did you already implement some UI for that?

[11:17] Currently the latter tasks are also a pain if the user wants to do it. Can only be done with macro right now.

[11:18] <Spiso> I didn't implement any UI yet, I would say that there could be a right click on formula giving option to align all, or just that formula to baseline

[11:18] Right now the preferred (or better minimum requirement) to activate the new alignment would be the one listed as b) above.

[11:18] <Spiso> and new formulas would be aligned new way

[11:18] <Fridrich> what I see in the two students is quite substantial work on the code, I would prefer not to start to add options for ui in this time. The ui is the last bit of it and not really fun in the timeframe of GSoc that is in its second half

[11:19] <Fridrich> we don't expect the stuff to be finished and polished at the GSoC end

[11:19] <Fridrich> but I would love not to have a load of half-started paths in it

[11:19] <Fridrich> instead, let us finalize one thing and then do more

[11:20] Fridrich: fine for me. I expected that the students wanted to be included in the discussion about the UI. But of course we can do that later.

[11:20] mba: do you agree with implementing the UI in the 'Object' dialog for now only?

[11:20] * ericb2 is back

[11:20] apologies for the delay

[11:20] <Fridrich> mba: if it is fun for them, why not :)

[11:21] <Fridrich> mba: I dislike things that are half-started and left unfinished and bitrotten later

[11:21] <Fridrich> mba: there, it is better not to start and have beer instead

[11:22] <Spiso> well, personally I don't need to work on the UI if you imagined doing it yourself :)

[11:22] <Fridrich> Spiso: no, you will work on it when you stick with the community after gsoc :)

[11:23] <Spiso> Fridrich: that's alright then, but this precisely gives half finished work, when I make some changes that can't be applied because of yet missing UI

[11:24] <Fridrich> Spiso: if it is really needed to have something functional, I am all for it

[11:25] Spiso: if everything you implement just contributes to the new alignment that later on will be activated through a yet undefined UI, I don't see a risk.

[11:25] If I can :  I tested the code for base alignment, and I'm not sure it works oon all cases.

[11:26] <Spiso> ok, and is there something needed to be done now, to start implementing it into openoffice?

[11:26]  thorsten tl13

[11:27] mba, tl13 : the current work in progress is : http://eric.bachard.free.fr/OOo4Kids/patches/July_2010/21st_july/complete_baseline_2010-7-13.diff

[11:27] A cws so that we can verify that the code compiles properly on all platform.

[11:27] tl13: no problem with that :)

[11:27] <Fridrich> ericb2: so those cases that don't work, we need to fix them :)

[11:27] Spiso: we "only" need to have a patch that applies to an existing DEV300 milestone that someone with commit access could put into a workspace ("CWS").

[11:27]  Fridrich froumi

[11:27] Fridrich: exactly

[11:28] <Spiso> ericb2: we can discuss that later, if you didn't want to include it into this meeting (I mean there is no point to do it here)

[11:28] <Fridrich> mba: ok, and SCA signed by the author :)

[11:28] mba: I really wanted to make a point with you. Firstly, because the backport is NOT a problem, but the technical side could be. and this is the poin

[11:28] point

[11:28] <Fridrich> but as I said, CWS I will take care of

[11:28] We (Thomas and I) can help with building and QAing the stuff.

[11:28] mba: thanks

[11:29] just looking at the diff it seems rather short in changing node.cxx. Thus I would agree that probably something is still missing here.

[11:29] <Spiso> tl13: node.cxx doesn't need much change

[11:29] <Spiso> tl13: why do you expect it?

[11:30] If I got a cws abd running office I'm willing to look a bit more into it and check if I find some node types that still need to be changed.

[11:30] BTW: I will be on vacation between July 29th and August 19th. And I have to leave in ~10 minutes as I have to go to a meeting.

[11:31] <Fridrich> ok

[11:31] <Spiso> mba: ok, thanks

[11:31] Spiso: first, solve the technical issues. Then we'll dive into the OOo QA and co process

[11:31] Because I would expect that at least the Arrange function of several nodes needed to be changed. On the other hand I have not yet looked at rect.cxx and maybe you put the relevant code there.

[11:31] Spiso: this is not a problem

[11:32] <Spiso> ericb2: ok, so do we want to commit it into CWS soon or not?

[11:32] Spiso: no need to commit before we are certain the issue is solved AND mba and tl agreed on the solution, or at least confirmed the possibility works

[11:33]  scorpius_ Spiso

[11:33] That is you want me (and maybe QA as well) to apply the patch first and do a little testing?

[11:33] tl13: please do

[11:34] ok. Thus I assume the alignment is already activated by default, right?

[11:34] tl13: if this can help, I think there is still one remaining issue when several formulas are in the same line. Probably a little change, but there is probable an improvement needed

[11:34] <Spiso> tl13: yes

[11:35] ericb2: hen and egg problem? Without a patch being applicable to an upstream workspace we can only review the code, but not test it. And without actually applying the existing patch and probably changing it we won't get a suitable patch for upstream OOo. ;-)

[11:35] <Fridrich> mba: you want a patch against dev300? I will do it for you :)

[11:35]  Fridrich froumi

[11:35] mba: we'll care to provide you an appliable .. Fridrich answered for me :)

[11:36] ok. But I will not do this right away. But lets say I'd like to start by the mid of next week and hopefully give some results by the end of it. Would that be ok?

[11:36] Fridrich: thank you, that's sufficient to work with it.

[11:36] <Fridrich> mba: just pray my wife does not give birth this week

[11:36] <Fridrich> she is in 38th week :)

[11:37] I see. *congrats*

[11:37] at all : the log will be online at : http://wiki.ooo4kids.org/index.php/ImproveMathEquationEditor/IRCMeetings/June2010/21st_July2010

[11:37] <Fridrich> then, it will be ericb2 who will port the patch if I am incapacitated :)

[11:37]  Fridrich froumi

[11:37]  Fridrich froumi

[11:37] Fridrich: no problem, I'll create a working patch for DEV300_m84 or so

[11:38] Then can we agree that I will get a patch based on DEV300_m83 (or higher) until mid of next week?

[11:38] <Fridrich> ericb2: m85 is out since tonight :)

[11:38] Sounds great!

[11:38] Fridrich: All the best for you and your family - fortunately there are events that are more important and more pleasing than code hacking. :-)

[11:39] mba: tl13 : for the record, dont forget all information is : http://wiki.ooo4kids.org/index.php/ImproveMathEquationEditor/Baseline_AlignmentEquations

[11:39] <Fridrich> mba: more plesing? another 6 months of sleepless nights for 6 minutes of pure pleasure :)

[11:39] <Fridrich> mba: just kidding :)

[11:39] :-)

[11:39] lol

[11:40] OK, then I expect to get the patch via email.

[11:40] mba tl13 : + http://wiki.ooo4kids.org/index.php/User:MSpisiak to follow the progress

[11:40] tl13: will do. Say in the two next days

[11:40] <Spiso> ericb2: oh hack, this one needs more attention

[11:40] As said about Tuesday next week will be fine

[11:41] No need to hurry ^__

[11:41] Excellent! Sorry guys, I have to leave now. I have to answer to call of duty.

[11:41] Spiso: I do a full build, on several OS, to be sure

[11:41] Then, are we done for now?

[11:41] <Fridrich> yup, I guess

[11:42] mba: thanks for attending :)

[11:42] ok. Thanks and good luck.

[11:42] See ya!

[11:42] <Spiso> thanks guys, see you too

[11:42] <Fridrich> thanks to all for attending