Feat/add webservice api by nhorlava · Pull Request #36 · mpilhlt/llamore

nhorlava · 2026-05-15T07:42:59Z

Closes #31

updated triggers for hf api

Copilot

Pull request overview

This PR exposes llamore as a FastAPI-based web service (closes #31). It adds a new api/ package with the FastAPI app, a Dockerfile and HuggingFace Spaces metadata, optional [api] extras in pyproject.toml/uv.lock, and CI jobs that deploy the app to a HF Space. It also makes a small fix in GeminiExtractor to delete uploaded PDFs and includes some unrelated notebook regeneration.

Changes:

New api/api.py exposing /extract/{openai,gemini}/{text,pdf} endpoints with API-key auth, request models, and PDF size validation.
Packaging and CI: optional api extra in pyproject.toml/uv.lock, plus deploy-api and rebuild-hf workflow jobs targeting a HF Space.
Minor: delete uploaded Gemini files after generation, .gitignore for notebooks/data, and regenerated quick_start.ipynb cells that include an unrelated schema change.

Reviewed changes

Copilot reviewed 7 out of 10 changed files in this pull request and generated 14 comments.

Show a summary per file

File	Description
api/api.py	New FastAPI service with auth, schemas, and extract endpoints.
api/Dockerfile	Image for HF Space; pins install to a feature branch.
api/README.md	HF Space front-matter.
pyproject.toml	Adds `[api]` optional dependency group.
uv.lock	Lock updates for FastAPI/uvicorn stack; also tightens lxml to `==5.3.1`.
.github/workflows/ci.yml	Adds `deploy-api` and `rebuild-hf` jobs and lxml system deps.
src/llamore/extractors.py	Deletes uploaded PDF from Gemini after `generate_content`.
notebooks/quick_start.ipynb	Regenerated outputs reflect a Person/Reference schema change not present in this PR; kernel metadata changed.
.gitignore	Excludes `notebooks/data`.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

+
+        if pdf:
+            self._client.files.delete(name=file.name)


+RUN pip install uv
+
+# RUN uv pip install --system git+https://github.com/mpilhlt/llamore.git[api]
+RUN uv pip install --system "git+https://github.com/mpilhlt/llamore.git@feat/add_webservice_api#egg=llamore[api]"


+RUN useradd -m -u 1000 user
+
+WORKDIR /app
+
+RUN mkdir -p logs
+
+RUN pip install uv
+
+# RUN uv pip install --system git+https://github.com/mpilhlt/llamore.git[api]
+RUN uv pip install --system "git+https://github.com/mpilhlt/llamore.git@feat/add_webservice_api#egg=llamore[api]"
+
+COPY --chown=user api.py .
+
+EXPOSE 7860
+
+CMD ["uvicorn", "api:app", "--host", "0.0.0.0", "--port", "7860"]


+              cd api
+              git init -b main
+              git config user.email "ci@github"
+              git config user.name "CI"
+              git remote add hf https://user:${HF_TOKEN}@huggingface.co/spaces/Llamore/api
+              git add .
+              if git diff --cached --quiet; then
+                echo "No changes to deploy — skipping push."
+                exit 0
+              fi
+              git commit -m "deploy from ${GITHUB_SHA}"
+              git push hf main --force


+    rebuild-hf:
+        runs-on: ubuntu-latest
+        environment: hf
+        needs: [tests, deploy-api]
+        if: |
+          always()
+          && (needs.tests.result == 'success'
+          || needs.deploy-api.result == 'success'
+          || github.event_name == 'workflow_dispatch')
+        steps:
+          - name: Trigger HF Space rebuild
+            env:
+              HF_TOKEN: ${{ secrets.HF_TOKEN }}
+            run: |
+              curl --fail -X POST \
+                "https://huggingface.co/api/spaces/Llamore/api/restart?factory=true" \
+                -H "Authorization: Bearer ${HF_TOKEN}"


+api_key_header = APIKeyHeader(name="X-Llamore-API-Key", scheme_name="Llamore API Key", auto_error=False)
+provider_key_header = APIKeyHeader(name="X-LLM-Provider-Key", scheme_name="LLM Provider Key", auto_error=False)


@@ -441,7 +492,7 @@
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
-   "version": "3.10.16"
+   "version": "3.9.20"


@@ -166,7 +166,7 @@
      "    \"Person\": {\n",
      "      \"description\": \"Contains a proper noun or proper-noun phrase referring to a person, possibly including one or more of the person's forenames, surnames, honorifics, added names, etc.\",\n",
      "      \"properties\": {\n",
-      "        \"forename\": {\n",
+      "        \"first_name\": {\n",
      "          \"anyOf\": [\n",
      "            {\n",
      "              \"type\": \"string\"\n",
@@ -176,8 +176,21 @@
      "            }\n",
      "          ],\n",
      "          \"default\": null,\n",
-      "          \"description\": \"Contains a forename, given or baptismal name.\",\n",
-      "          \"title\": \"Forename\"\n",
+      "          \"description\": \"Contains a first name, given or baptismal name.\",\n",
+      "          \"title\": \"First Name\"\n",
+      "        },\n",
+      "        \"middle_name\": {\n",
+      "          \"anyOf\": [\n",
+      "            {\n",
+      "              \"type\": \"string\"\n",
+      "            },\n",
+      "            {\n",
+      "              \"type\": \"null\"\n",
+      "            }\n",
+      "          ],\n",
+      "          \"default\": null,\n",
+      "          \"description\": \"Contains a middle name, written between a person's first and surname. It is often abbreviated.\",\n",
+      "          \"title\": \"Middle Name\"\n",
      "        },\n",
      "        \"surname\": {\n",
      "          \"anyOf\": [\n",
@@ -191,6 +204,32 @@
      "          \"default\": null,\n",
      "          \"description\": \"Contains a family (inherited) name of a person, as opposed to a given, baptismal, or nick name.\",\n",
      "          \"title\": \"Surname\"\n",
+      "        },\n",
+      "        \"name_link\": {\n",
+      "          \"anyOf\": [\n",
+      "            {\n",
+      "              \"type\": \"string\"\n",
+      "            },\n",
+      "            {\n",
+      "              \"type\": \"null\"\n",
+      "            }\n",
+      "          ],\n",
+      "          \"default\": null,\n",
+      "          \"description\": \"Contains a connecting phrase or link used within a name but not regarded as part of it, such as 'van der' or 'of'.\",\n",
+      "          \"title\": \"Name Link\"\n",
+      "        },\n",
+      "        \"role_name\": {\n",
+      "          \"anyOf\": [\n",
+      "            {\n",
+      "              \"type\": \"string\"\n",
+      "            },\n",
+      "            {\n",
+      "              \"type\": \"null\"\n",
+      "            }\n",
+      "          ],\n",
+      "          \"default\": null,\n",
+      "          \"description\": \"Contains a name component which indicates that the referent has a particular role or position in society, such as an official title or rank.\",\n",
+      "          \"title\": \"Role Name\"\n",


+import logging
+import os
+import secrets
+import tempfile
+import traceback
+from contextlib import asynccontextmanager
+from pathlib import Path
+from typing import Annotated, Any, Dict, List, Literal, Optional
+
+from fastapi import Depends, FastAPI, Form, HTTPException, Request, Security, UploadFile
+from fastapi.responses import JSONResponse
+from fastapi.security import APIKeyHeader
+
+from llamore import (
+    GeminiExtractor,
+    LineByLinePrompter,
+    OpenaiExtractor,
+    References,
+    SchemaPrompter,
+)
+from pydantic import BaseModel, BeforeValidator, Field
+
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
+    handlers=[logging.StreamHandler()],
+)
+logger = logging.getLogger(__name__)
+
+


+    text: str = Field(..., min_length=1, description="Raw text to extract references from.")
+
+
+
+class GeminiExtractTextRequest(GeminiExtractionConfig):
+    """Request body for Gemini text extraction."""
+
+    text: str = Field(..., min_length=1, description="Raw text to extract references from.")
+
+


dcfidalgo and others added 20 commits November 26, 2025 11:00

fix forename in quick_start notebook

5c983fc

delete pdf file from cloud in case of gemini extractor

399d94c

add evaluate notebook

a503fe5

darft of fastapi

5492ac0

api with actions - test

07a8c10

merge from main

7296646

upd lock

1bf7d99

chore: use uv.lock from main

14ada8a

fix dependancies

2bdc89c

fix dependancies

ce7a89e

fix

93b7af4

fix api push to hf

672b6a8

Update ci.yml

95d9f46

updated triggers for hf api

cleaner requests

213fc22

cleaner requests

c0e037c

separate clients

177f6e4

add workflow_dispatch to deploy-api

8d005b4

fix deploy-api

59700c9

separate clients

71bd10a

refactoring

29315ab

nhorlava self-assigned this May 15, 2026

Copilot AI review requested due to automatic review settings May 15, 2026 07:43

Copilot started reviewing on behalf of nhorlava May 15, 2026 07:43 View session

nhorlava deployed to hf May 15, 2026 07:43 — with GitHub Actions Active

Copilot AI reviewed May 15, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Feat/add webservice api#36

Feat/add webservice api#36
nhorlava wants to merge 20 commits into
mainfrom
feat/add_webservice_api

nhorlava commented May 15, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

		api_key_header = APIKeyHeader(name="X-Llamore-API-Key", scheme_name="Llamore API Key", auto_error=False)
		provider_key_header = APIKeyHeader(name="X-LLM-Provider-Key", scheme_name="LLM Provider Key", auto_error=False)

Conversation

nhorlava commented May 15, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants