feat: convenience unstructured-get-json.sh update #3971
cragwolfe
merged
37 commits into
main
from
crag/unstructured-get-json-update
feat: Add VLM strategy support to unstructured-get-json.sh script
632aab6d
feat: Add --write-html option to unstructured-get-json with env var s…
1093ffee
feat: Add HTML generation using elements_json_to_html.py in unstructu…
f3c03fd3
chore: Replace python with python3 in shell script
5424613a
refactor: Replace direct function call with script invocation in unst…
e8d24aaa
refactor: Add elements without page number to 'none' page instead of …
6bfcb2f4
feat: Add support for --open-html option to automatically open HTML o…
4170a150
refactor: Remove echo statement when opening HTML file in browser
c0687e43
feat: Add support for --VLM-PROVIDER parameter in unstructured-get-js…
3fe657c9
feat: Add support for --vlm-model parameter in unstructured-get-json.sh
c455c92a
docs: Add documentation link for VLM strategy model types
52694518
style: Break long help text lines for improved readability
d10d0a0f
style: Improve formatting of --open-html help text line
23aa81c3
docs: Clarify $UNST_AUTO_OPEN_HTML should be set to true in help text
b810747c
style: Remove $ before env var in help text for clarity
b39d4d16
feat: Enhance VLM output filename with provider and model details
bf6aac43
fix: Handle elements with missing parent IDs by moving them to root l…
9e66fd5e
ignore aider files
f963b01d
feat: Add strategy conflict check for mutually exclusive options
f237daf6
refactor: Move strategy conflict check after argument processing
ebd833b7
fix: Add debug echo statements in unstructured-get-json.sh script
63a44f35
refactor: Remove debug echo statements from strategy conflict check
2e8f2866
fix: Remove debug echo statements from strategy conflict check function
7c02deea
fix: Remove debug echo statements in strategy conflict check function
bb88508c
refactor: Remove strategy conflict function and replace with inline c…
81d04f80
doc
62b13983
docs: Update help text for VLM strategy options in unstructured-get-j…
a280b7d0
feat: Add validation for VLM provider and model without VLM strategy
0e23bf16
refactor: Improve HTML generation by using jq to check and create HTM…
ac72c860
fix: Update script to use .metadata.text_as_html instead of .text_as_…
62c9d784
fix: Adjust jq expression for checking metadata.text_as_html field
98b3df56
refactor: Improve jq expression for checking metadata.text_as_html field
80fc64e3
fix: Improve jq error handling for text_as_html metadata field
c7836e5d
revert convert.py change
bd08722f
shfmt
87facd48
cragwolfe
changed the title
bump version
21e40c9c
ryannikolaidis
approved these changes
on 2025-03-29
replace the jq check with checking if in vlm mode, since they are one…
012f113c
ryannikolaidis
approved these changes
on 2025-03-31
Login to write a write a comment.
Login via GitHub