Move parsing of TRYSERVER_* so options.root is not overwritten

Move parsing of TRYSERVER_* so options.root is not overwritten

Since the helper function for the parsing of the TRYSERVER_* variables relies on them being unset (i.e. value is None) in order to apply the settings from codereview.settings, this parsing must take place before the code that figures out the root path (since that sets the options.root value if it's not set). 
Otherwise the TRYSERVER_ROOT setting will never be set, even if configured in codereview.settings, essentially making it useless.

This has probably not been discovered previously since almost all Chrome projects use src as the root dir. In WebRTC and some other projects we use 'trunk' usually, which is giving us trouble in our own try server  setups (where we have to use 'src' because many other scripts have that root dir hardcoded).

I am not sure if there is a case for another project where this change may have any negative effect. Please let me know if theres's any investigation needed that I can assist with.

BUG=none
TEST=submitting try job to local try master with a codereview.settings file with TRYSERVER_ROOT set.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=173694

[M-A Ruel, 2012-12-13 23:58:06]

https://codereview.chromium.org/11574007/diff/1/trychange.py
File trychange.py (right):

https://codereview.chromium.org/11574007/diff/1/trychange.py#newcode171
trychange.py:171: self._GclStyleSettings()
I'm pretty sure there was a reason for it to be done in the other order.

But I can't recall.

lgtm, just be prepared to revert in case of tool breakage reports.

[kjellander_chromium, 2012-12-18 08:52:15]

I'll go ahead and commit it today. I'll be prepared to revert it but I assume
you guys are the ones that will be notified first since I'm in STO.

https://codereview.chromium.org/11574007/diff/1/trychange.py
File trychange.py (right):

https://codereview.chromium.org/11574007/diff/1/trychange.py#newcode171
trychange.py:171: self._GclStyleSettings()
I think it's pretty safe to move it. 
All the code below that was added before the parsing was moved there in
https://codereview.chromium.org/8174009
I guess since no on seems to use TRYSERVER_ROOT this problem was not noticed.


On 2012/12/13 23:58:06, Marc-Antoine Ruel wrote:
> I'm pretty sure there was a reason for it to be done in the other order.
> 
> But I can't recall.
> 
> lgtm, just be prepared to revert in case of tool breakage reports.

[phoglund_chromium, 2012-12-18 09:03:14]

I think the change is correct. The two methods only have interactions on
options.root in common as far as I can see, and that interaction seems to be
corrected by this patch. I don't think there are other side effects that would
be affected by the change in call order.

[commit-bot: I haz the power, 2012-12-18 09:28:43]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kjellander@chromium.org/11574007/1

[commit-bot: I haz the power, 2012-12-18 09:31:15]

Change committed as 173694