Open Bug 63766 Opened 24 years ago Updated 2 years ago

clean up messy include structure in layout

Categories

(Core :: Layout, defect, P4)

x86
Linux
defect

Tracking

()

Future

People

(Reporter: dbaron, Unassigned)

References

Details

Attachments

(1 file, 3 obsolete files)

In layout, there are a number of files exported from src directories, and some
include paths that are bigger than they need to (or ought to) be.  If we're ever
going to successfully split layout, this needs to be cleaned up.

I did a bit of work on this this afternoon, and I'll attach a patch that removes
a bunch of exports and shrinks a few include paths.  (The gcc-generated .deps
directory is great for seeing how the include paths are used.)
And I already noticed a rather serious typo...  Well, you get the idea, anyway.
r=jst for the last version of the patch (apart from the unrelated changes to
layout/xul/base/src/nsBoxToBlockAdaptor.cpp).
I just finished a distclean and delete-the-whole-objdir build (as close as one
can get to a fresh tree without a new checkout, I think), and it built and ran fine.

And, yes, I didn't mean to include those changes in nsBoxToBlockAdaptor.cpp in
the patch.
nice work.  sr=buster.

one minor item:
===================================================================
RCS file: /cvsroot/mozilla/layout/build/makefile.win,v
retrieving revision 3.42
diff -u -r3.42 makefile.win
--- makefile.win	2000/11/18 02:13:34	3.42
+++ makefile.win	2000/12/27 18:22:51
@@ -52,6 +52,7 @@

 LINCS=-I$(PUBLIC)\xpcom -I$(PUBLIC)\raptor -I$(PUBLIC)\dom \
 	-I$(PUBLIC)\js -I..\..\style\src -I..\html\base\src -I..\base\src \
+
-I..\html\style\src \
 	-I..\xul\content\src \
 	-I..\xul\base\src \

can we remove -I..\..\style\src  ?


on a big system checkin like this, here's what I think we need to do:
1) get somebody to apply the patch to their tree, one person per major platform,
and verify that it builds ok.  sounds like you've already done this for linux.
2) notify the other platform owners that the makefiles have changed (by posting
to .builds, maybe.)
3) get in contact with folks who have major #ifdef blocks of code and give them
a heads-up.  mathml, ibm bi-di, etc.  maybe just a posting to .layout
4) give (2) and (3) a couple of days to settle in, address any comments or
concerns, then carpool in the changes.

I really don't think the changes are that risky or complicated.  I'd been hoping
to check them in sooner.
Buster, I do agree with you that a checkin like this one would be good to get
tested on many platforms but given that it's saturday morning and it's the day
before new-years and not that many people are working I think it's acceptable to
just land this as is now. I just chatted with dbaron and we'll give it a shot,
I'll help dealing with the possible redness.
Status: NEW → ASSIGNED
Priority: -- → P3
Target Milestone: --- → mozilla0.8
Target Milestone: mozilla0.8 → mozilla0.9
Is this bug still alive? Shouldn't the patch2 changed after the layout content
split. I would volunteer to test a patch under win98.
Reality check.  Moving out to 0.9.1.
Target Milestone: mozilla0.9 → mozilla0.9.1
Target Milestone: mozilla0.9.1 → mozilla0.9.2
Target Milestone: mozilla0.9.2 → Future
Attachment #21316 - Attachment is obsolete: true
Attachment #21321 - Attachment is obsolete: true
Attachment #21333 - Attachment is obsolete: true
Attachment #21511 - Flags: needs-work+
QA Contact: chrispetersen → layout
Assignee: dbaron → nobody
Status: ASSIGNED → NEW
Severity: normal → S3
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: