142282 - Add Supercede attribute to component sections in config.ini

Status: VERIFIED FIXED

(SeaMonkey :: Installer, defect)

Description

[dprice (gone)]

We want to be able to force the installer to skip a component if a condition is
met, like a file existing, a registry key being set, or a newer version is
installed.  To do this we can add the following optional attributes to component
sections in the config.ini

SupercedeType=[File Exists/Registry Key/Version]
SupercedeFile=Path to file
SupercedeKey=Key Name
SupercedeVersion=number

Comment 1

[dprice (gone)]

Attachment: patch .1

This patch isn't perfect.  It greys out and unselects the right item.  But full
and recommended installs are bringing up the additional componments window. 
I'm not sure why, so I'm posting it here in case someone has a suggestion.

Comment 2

[Jaime Rodriguez, Jr.]

Needed to support buscape 14738
http://bugscape.netscape.com/show_bug.cgi?id=14738

Comment 3

[dprice (gone)]

Attachment: patch

Thanks to some great help from ssu, this patch it working.  It only covers the
File Exists Supercede type.  It does all that's needed for bugscape 14738

Comment 4

[Jaime Rodriguez, Jr.]

Who can provide r/sr= for this fix?

Comment 5

[timeless]

can someone suggest a non commercial example of how this would be used?

Comment 6

[dprice (gone)]

Netscape ships with Real Player 8.  But Real One is out.  If Real One is on the
system, the Real Player 8 installer simply quits without doing anything.  In
that case Downloading the RP8 installer is a waste of time.  With the Supercede
attribute we can force the installer to skip the RP8 download if Real One is
found on the system.

Comment 7

[Sean Su]

Comment on attachment 82381 [details] [diff] [review]
patch

sorry David.  the patch still needs work.  A new feature that Curt added to the
installer (a while back) contained bugs that this feature exposed.  I had to
fix those bugs in order to get your patch to work.

I'll attach the patch that works for me.

David, please do a diff between your patch and my patch to see the changes I
made.

Comment 8

[Sean Su]

Attachment: patch v1.0 (dprice w/minor ssu fixes)

David, sorry for creating my own patch ontop of yours.	I'm just trying to get
this bug fixed asap due to the hectic schedule that we have.

Could you also make sure that this new patch looks okay and still works as
expected?  Thanks.

Comment 9

[Syd Logan]

I'll work with sean on getting a review

Comment 10

[dprice (gone)]

r=dprice for ssu's changes to my fix.  

Comment 11

[Syd Logan]

Comment on attachment 82451 [details] [diff] [review]
patch v1.0 (dprice w/minor ssu fixes)

r=syd

Comment 12

[Scott MacGregor]

Comment on attachment 82451 [details] [diff] [review]
patch v1.0 (dprice w/minor ssu fixes)

sr=mscott...sean went over the patch with me.

Comment 13

[Jaime Rodriguez, Jr.]

adt1.0.0+ (on ADT's behalf) approval for checkin to the 1.0 brnach. Pls check
this in after you receive a= from Drivers, then add the fixed1.0.0 keyword.

Comment 14

[Syd Logan]

dprice, please get this in the trunk and mark it fixed

Comment 15

[chris hofmann]

a=chofmann for the 1.0 branch.  get this in and let's shake on it hard in the
5/7 builds.  let's get testing from both the installer and plugin's perspective
on this one.

Comment 16

[chris hofmann]

Comment on attachment 82451 [details] [diff] [review]
patch v1.0 (dprice w/minor ssu fixes)

a=chofmann for the 1.0 branch

Comment 17

[Gayatri Rimola]

Set QA contact to K'trina.

Comment 18

[dprice (gone)]

on the branch or the trunk

Comment 19

[dprice (gone)]

branch or trunk... guess which?  I mean on the branch AND the trunk 

Comment 20

[K'Trina Medina]

Verified code fix