• Joined on 2021-11-20
ungleich-gitea commented on issue ungleich-public/cdist#80 2021-11-20 15:35:44 +00:00
cycle detection in object dependencies broken?

changed the description

ungleich-gitea commented on issue ungleich-public/cdist#80 2021-11-20 15:35:42 +00:00
cycle detection in object dependencies broken?

if first type in initial manifest doesn't have CDIST_ORDER_DEPENDENCY, then it works:

$ for f in manifest/bug type/__foo/manifest type/__bar/manifest type/__baz/manifest; do echo; while read -r…
ungleich-gitea commented on issue ungleich-public/cdist#80 2021-11-20 15:35:40 +00:00
cycle detection in object dependencies broken?

@ander CDIST_ORDER_DEPENDENCY instructs cdist to record last created type as a requirement for current processed type, from type creation order list.

For the first example above, this is what…

ungleich-gitea commented on issue ungleich-public/cdist#80 2021-11-20 15:35:38 +00:00
cycle detection in object dependencies broken?

@steven @nico What if we try to change how CDIST_ORDER_DEPENDENCY works? How about defining context in which this env var is valid, and a local list of type creation order start from beginning in…

ungleich-gitea commented on issue ungleich-public/cdist#80 2021-11-20 15:35:37 +00:00
cycle detection in object dependencies broken?

CDIST_ORDER_DEPENDENCY is evil, it breaks things.

yes, but it is necessary evil. i, for one, support this evil env var.

ungleich-gitea commented on issue ungleich-public/cdist#80 2021-11-20 15:35:36 +00:00
cycle detection in object dependencies broken?

CDIST_ORDER_DEPENDENCY should only be valid within the manifest where it is used. That we have a global file tracking this is at least one of the problems.

What if we have a 'type_creation_order'…

ungleich-gitea commented on issue ungleich-public/cdist#80 2021-11-20 15:35:34 +00:00
cycle detection in object dependencies broken?

@steven Yes, that is in the idea I am talking about. I will experiment with it.

ungleich-gitea commented on issue ungleich-public/cdist#80 2021-11-20 15:35:33 +00:00
cycle detection in object dependencies broken?

mentioned in merge request !815

ungleich-gitea commented on issue ungleich-public/cdist#80 2021-11-20 15:35:30 +00:00
cycle detection in object dependencies broken?

closed

ungleich-gitea commented on issue ungleich-public/cdist#79 2021-11-20 15:35:29 +00:00
__clean_path shellcheck issue

changed the description

ungleich-gitea commented on issue ungleich-public/cdist#79 2021-11-20 15:35:28 +00:00
__clean_path shellcheck issue

we have "stdin inside a loop" issue here if i use while read.

so, __file foo < /dev/null or # shellcheck disable=SC2013 ?

ungleich-gitea commented on issue ungleich-public/cdist#79 2021-11-20 15:35:26 +00:00
__clean_path shellcheck issue

@ander I wouldn't ignore and disable this. What if file contains spaces? Current loop wouldn't work.

ungleich-gitea commented on issue ungleich-public/cdist#79 2021-11-20 15:35:25 +00:00
__clean_path shellcheck issue

you are right

ungleich-gitea commented on issue ungleich-public/cdist#79 2021-11-20 15:35:23 +00:00
__clean_path shellcheck issue

mentioned in merge request !817

ungleich-gitea commented on issue ungleich-public/cdist#79 2021-11-20 15:35:22 +00:00
__clean_path shellcheck issue

closing because of spaces

ungleich-gitea commented on issue ungleich-public/cdist#79 2021-11-20 15:35:21 +00:00
__clean_path shellcheck issue

closed

ungleich-gitea commented on issue ungleich-public/cdist#78 2021-11-20 15:35:19 +00:00
__file and __directoy broken if path contains space

cdist does not support file names containing space. It would also break require="" variable. Through all the levels of shell and writing to files and reading from files it would be complicated.

So,…