Luke Hsiao
Luke Hsiao
> IMO it's misleading to add a marker file that says "yes, this library supports type checking", but the library doesn't support type checking (yet). This is fair, but again...
``` ❯ git-cliff --context --unreleased --bump | jaq . [ { "version": "0.1.10", "commits": [ { "id": "15f5037f5bacb914b41514b33af4abc409515a2e", "message": "remove extra `v` from version links", "body": null, "footers": [], "group":...
Another note, this appears to be fixed in `main`, so perhaps all we need is a new release? ## On main ``` ❯ ../../orhun/git-cliff/target/release/git-cliff --unreleased --tag v0.2.2 --context | jaq...
As another reference, semantic-release is a popular way to manage releases from conventional commits, and their workflow is described here: https://semantic-release.gitbook.io/semantic-release/recipes/release-workflow/maintenance-releases
FWIW I use fish with selenized in my terminal and it works great.
The result of the sorting seems inconsistent [in TreeExtract.py](https://github.com/HazyResearch/pdftotree/blob/fbe974aed8411c966ad162499d1c934a7fd06001/pdftotree/TreeExtract.py#L163) between Python2 and Python3. ```py boxes.sort(key=cmp_to_key(two_column_paper_order)) ``` ```py def two_column_paper_order(b1, b2): ''' b1 = [b1.type, b1.top, b1.left, b1.bottom, b1.right] b2 =...
It looks like `two_column_paper_order` is inconsistent. ```py from pprint import pprint from functools import cmp_to_key boxes = [['header', 52.92960000000005, 59.8111, 71.16359999999997, 286.05310000000003], ['section_header', 72.88620000000003, 59.8111, 91.12019999999995, 163.90329999999997], ['section_header', 111.77690000000007, 59.8111,...
We've boiled this down for a need to implement a function to detect the number of columns in a document, and then order the text between the columns in reading...
@HodeiG I'm sorry, I'm afraid I no longer have write access to this repository, consequently, I have no way to help maintain it even if I wanted to. If you're...
I see your point. One concern I would have is that the test data can be quite large. I'm not sure we actually *want* to bundle that together. Is there...