skeema / skeema / 5263273821
92%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 73
Run time 1s
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 93.233% (+0.02%) from 93.209%
5263273821

push

github

evanelias 
init,pull: retain object name casing for .sql files

Skeema v1.10.0 unintentionally introduced a change in how `skeema init` and
`skeema pull` interact with mixed-case object names on MacOS and Windows,
when creating new .sql files for objects. Prior to v1.10.0, an object's name
would be used as-is for the .sql file name; however v1.10.0 would use a
downcased version of the object name for naming the .sql file on Mac/Win.

This commit restores the previous behavior of using an object's mixed-case
name as-is when creating and naming a new .sql file.

Regardless of Skeema version, users should note that this logic only
controls the *default* naming for *new* files created by `skeema init` and
`skeema pull`. Once a .sql file has been created, users can freely rename the
file in any way (not just casing) and/or move SQL statements between files, or
combine multiple statements into a single file. Skeema does NOT require that
statements remain in files using their default file names. As long as the
statements are contained in one or more files with a .sql extension in the
same *directory*, all Skeema commands will still work properly, and `skeema
pull` can even dynamically update statements regardless of location.

This commit also fixes some edge cases involving filesystem case sensitivity/
insensitivity which existed in v1.9.0 and previous: on a case-insensitive
filesystem (Mac or Windows), if some object names differ only by casing (e.g.
`Foo` and `foo` both exist in the database), `skeema init` and `skeema pull`
had subtle bugs in their logic to dump objects to the filesystem. Meanwhile
on a case-sensitive filesystem (Linux), this would "correctly" create separate
files Foo.sql and foo.sql; however, this would prove to be problematic if
also using the same schema repo on case-insensitive systems. As of this
commit, Skeema will avoid introducing such naming conflicts, by now
preferring to place `Foo` and `foo` in the same .sql file. That said, if a
given... (continued)

9025 of 9680 relevant lines covered (93.23%)

1.11 hits per line

Jobs
ID Job ID Ran Files Coverage
1 5263273821.1 0
93.23
 GitHub Action Run
Source Files on build 5263273821
Detailed source file information is not available for this build.