1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
|
1
2
3
4
5
6
7
8
9
10
11
|
-
-
-
-
-
-
-
-
-
-
-
-
|
An informal list of FIXMEs which don't warrant full tickets...
This is not *intended* to be a list of long-lived fixme's, but more of a scratchpad for things which come up spontaneously and shouldn't be forgotten.
# f-ci (checkin)
```
$ ls
abc def ghi
$ f-ci -m ... abc ghi xyz
```
It "really should" report files which it does not know about as an error but it currently silently skips them. Currently the only way we have to report that a given filename passed to it was not picked up by `fsl_checkin_enqueue()` is to check the size of `fsl_cx::ckin.selectedIds` (a `fsl_id_bag`) before and after calling `fsl_checkin_enqueue()`. However, since that function accepts a directory name and a dir name with no modified files is indistinguishable from a directory we know nothing about (without further tests only to check for this), checking the size of `fsl_cx::ckin.selectedIds` is insufficient. Ergo, we don't currently have a way to reliably detect the difference without jumping through as-yet-undetermined hoops which would be created solely to catch that case.
One idea to resolve this would be to stat() each argument and ensure that it's either a file or a directory, but that would fail for files which are being removed as part of the commit and were removed from the filesystem by the user. Though we could catch such cases by checking the vfile table if the stat fails, it still leaves us in the unpleasant position of building up a lot of code simply for the sake of being able to report this usage error.
|