From c1ea215a574d6385b0497365ed0f113c7ee0f5a3 Mon Sep 17 00:00:00 2001 From: CyMule Date: Wed, 7 Jan 2026 10:36:50 -0500 Subject: [PATCH 1/2] perf: upgrade pdfminer-six to 20260107 Fixes ~15-18% performance regression introduced in 20251230 where f-strings were evaluated eagerly even when logging was disabled. See: https://github.com/pdfminer/pdfminer.six/issues/1233 Fix: https://github.com/pdfminer/pdfminer.six/pull/1234 --- CHANGELOG.md | 3 +++ prepline_general/api/__version__.py | 2 +- requirements/base.txt | 3 ++- requirements/constraints.in | 4 +++- 4 files changed, 9 insertions(+), 3 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 8529bfc6f..7482c1e6a 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,3 +1,6 @@ +## 0.0.92 +* Upgrade pdfminer-six to 20260107 to fix ~15-18% performance regression from eager f-string evaluation + ## 0.0.91 * Upgrade packages to resolve CVEs diff --git a/prepline_general/api/__version__.py b/prepline_general/api/__version__.py index 7cd18c1c9..9666db13e 100644 --- a/prepline_general/api/__version__.py +++ b/prepline_general/api/__version__.py @@ -1 +1 @@ -__version__ = "0.0.91" # pragma: no cover +__version__ = "0.0.92" # pragma: no cover diff --git a/requirements/base.txt b/requirements/base.txt index c1329f01e..8ee5bdfdb 100644 --- a/requirements/base.txt +++ b/requirements/base.txt @@ -216,8 +216,9 @@ pandas==2.3.3 # unstructured-inference pdf2image==1.17.0 # via unstructured -pdfminer-six==20251230 +pdfminer-six==20260107 # via + # -c requirements/constraints.in # unstructured # unstructured-inference pi-heif==1.1.1 diff --git a/requirements/constraints.in b/requirements/constraints.in index f213ea100..290080d80 100644 --- a/requirements/constraints.in +++ b/requirements/constraints.in @@ -5,4 +5,6 @@ #################################################################################################### numpy<2.0.0 # later versions of Starlette break middleware -starlette==0.41.2 \ No newline at end of file +starlette==0.41.2 +# pdfminer.six 20260107 includes performance fix +pdfminer-six==20260107 From cd8a7ef3f29ba69842972ff665ac26a05d307fb9 Mon Sep 17 00:00:00 2001 From: CyMule Date: Wed, 7 Jan 2026 11:04:37 -0500 Subject: [PATCH 2/2] bump preprocessing family yml --- preprocessing-pipeline-family.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/preprocessing-pipeline-family.yaml b/preprocessing-pipeline-family.yaml index 7b5199b7e..ad2c506a7 100644 --- a/preprocessing-pipeline-family.yaml +++ b/preprocessing-pipeline-family.yaml @@ -1,2 +1,2 @@ name: general -version: 0.0.91 +version: 0.0.92