deepresearch-flow 0.6.0__py3-none-any.whl → 0.6.1__py3-none-any.whl

deepresearch_flow/paper/db.py

@@ -31,7 +31,8 @@ from deepresearch_flow.paper.template_registry import (
 from deepresearch_flow.paper.render import resolve_render_template, render_papers
 
 try:
-    from pybtex.database import parse_file
+    from pybtex.database import BibliographyData, parse_file
+    from pybtex.database.output.bibtex import Writer
     PYBTEX_AVAILABLE = True
 except ImportError:
     PYBTEX_AVAILABLE = False
@@ -1015,11 +1016,18 @@ def register_db_commands(db_group: click.Group) -> None:
     def merge_group() -> None:
         """Merge paper JSON inputs."""
 
-    def _summarize_merge(output_path: Path, merged: list[dict[str, Any]], *, input_count: int) -> None:
+    def _summarize_merge(output_path: Path, merged: Any, *, input_count: int) -> None:
+        items: list[dict[str, Any]] = []
+        if isinstance(merged, dict):
+            raw_items = merged.get("papers")
+            if isinstance(raw_items, list):
+                items = [item for item in raw_items if isinstance(item, dict)]
+        elif isinstance(merged, list):
+            items = [item for item in merged if isinstance(item, dict)]
+
         field_set: set[str] = set()
-        for item in merged:
-            if isinstance(item, dict):
-                field_set.update(item.keys())
+        for item in items:
+            field_set.update(item.keys())
         field_list = sorted(field_set)
 
         console = Console()
@@ -1027,7 +1035,7 @@ def register_db_commands(db_group: click.Group) -> None:
         summary.add_column("Metric", style="bold")
         summary.add_column("Value")
         summary.add_row("Inputs", str(input_count))
-        summary.add_row("Items", str(len(merged)))
+        summary.add_row("Items", str(len(items)))
         summary.add_row("Fields", str(len(field_list)))
         summary.add_row("Output", str(output_path))
         console.print(summary)
@@ -1039,17 +1047,65 @@ def register_db_commands(db_group: click.Group) -> None:
                 field_table.add_row(name)
             console.print(field_table)
 
+    def _bibtex_entry_score(entry: Any) -> int:
+        fields = getattr(entry, "fields", {}) or {}
+        persons = getattr(entry, "persons", {}) or {}
+        person_count = sum(len(people) for people in persons.values())
+        return len(fields) + len(persons) + person_count
+
+    def _summarize_bibtex_merge(output_path: Path, *, input_count: int, entry_count: int, duplicate_count: int) -> None:
+        summary = Table(title="BibTeX Merge Summary")
+        summary.add_column("Metric", style="bold")
+        summary.add_column("Value")
+        summary.add_row("Inputs", str(input_count))
+        summary.add_row("Entries", str(entry_count))
+        summary.add_row("Duplicates", str(duplicate_count))
+        summary.add_row("Output", str(output_path))
+        Console().print(summary)
+
     @merge_group.command("library")
     @click.option("-i", "--inputs", "input_paths", multiple=True, required=True, help="Input JSON files")
+    @click.option("--template-tag", "template_tag", default=None, help="Template tag for merged output")
     @click.option("-o", "--output", "output_path", required=True, help="Output JSON file path")
-    def merge_library(input_paths: Iterable[str], output_path: str) -> None:
+    def merge_library(input_paths: Iterable[str], template_tag: str | None, output_path: str) -> None:
         paths = [Path(path) for path in input_paths]
         merged: list[dict[str, Any]] = []
+        tag_candidates: list[str] = []
         for path in paths:
-            merged.extend(load_json(path))
+            payload = load_json(path)
+            if isinstance(payload, dict):
+                tag = str(payload.get("template_tag") or "")
+                if tag:
+                    tag_candidates.append(tag)
+                papers = payload.get("papers")
+                if isinstance(papers, list):
+                    merged.extend(papers)
+                else:
+                    raise click.ClickException("Input JSON must be a list or {template_tag, papers}")
+            elif isinstance(payload, list):
+                merged.extend(payload)
+            else:
+                raise click.ClickException("Input JSON must be a list or {template_tag, papers}")
+        if not template_tag:
+            inferred = ""
+            for paper in merged:
+                if not isinstance(paper, dict):
+                    continue
+                inferred = str(paper.get("prompt_template") or paper.get("template_tag") or "")
+                if inferred:
+                    break
+            if inferred:
+                template_tag = inferred
+        if tag_candidates and not template_tag:
+            template_tag = tag_candidates[0]
+        if not template_tag:
+            template_tag = "unknown"
+        if tag_candidates and any(tag != template_tag for tag in tag_candidates):
+            click.echo("Warning: multiple template_tag values detected in inputs; using first")
         output = Path(output_path)
-        write_json(output, merged)
-        _summarize_merge(output, merged, input_count=len(paths))
+        bundle = {"template_tag": template_tag, "papers": merged}
+        write_json(output, bundle)
+        _summarize_merge(output, bundle, input_count=len(paths))
 
     @merge_group.command("templates")
     @click.option("-i", "--inputs", "input_paths", multiple=True, required=True, help="Input JSON files")
@@ -1201,6 +1257,62 @@ def register_db_commands(db_group: click.Group) -> None:
                 sample_table.add_row(*row)
             Console().print(sample_table)
 
+    @merge_group.command("bibtex")
+    @click.option("-i", "--input", "input_paths", multiple=True, required=True, help="Input BibTeX file paths")
+    @click.option("-o", "--output", "output_path", required=True, help="Output BibTeX file path")
+    def merge_bibtex(input_paths: Iterable[str], output_path: str) -> None:
+        if not PYBTEX_AVAILABLE:
+            raise click.ClickException("pybtex is required for merge bibtex")
+
+        paths = [Path(path) for path in input_paths]
+        if not paths:
+            raise click.ClickException("No BibTeX inputs provided")
+
+        for path in paths:
+            if not path.is_file():
+                raise click.ClickException(f"BibTeX file not found: {path}")
+
+        merged_entries: dict[str, tuple[Any, int]] = {}
+        duplicate_keys: list[str] = []
+        duplicate_seen: set[str] = set()
+
+        for path in paths:
+            bib_data = parse_file(str(path))
+            for key, entry in bib_data.entries.items():
+                score = _bibtex_entry_score(entry)
+                if key not in merged_entries:
+                    merged_entries[key] = (entry, score)
+                    continue
+                if key not in duplicate_seen:
+                    duplicate_seen.add(key)
+                    duplicate_keys.append(key)
+                _, existing_score = merged_entries[key]
+                if score > existing_score:
+                    merged_entries[key] = (entry, score)
+
+        output = Path(output_path)
+        output.parent.mkdir(parents=True, exist_ok=True)
+        out_data = BibliographyData()
+        for key, (entry, _) in merged_entries.items():
+            out_data.entries[key] = entry
+        with output.open("w", encoding="utf-8") as handle:
+            Writer().write_stream(out_data, handle)
+
+        _summarize_bibtex_merge(
+            output,
+            input_count=len(paths),
+            entry_count=len(merged_entries),
+            duplicate_count=len(duplicate_keys),
+        )
+
+        if duplicate_keys:
+            preview_limit = 20
+            preview = ", ".join(duplicate_keys[:preview_limit])
+            if len(duplicate_keys) > preview_limit:
+                preview = f"{preview}, ... (+{len(duplicate_keys) - preview_limit} more)"
+            note = "Kept entry with most fields; ties keep first input order."
+            Console().print(Panel(f"{note}\n{preview}", title=f"Duplicate keys ({len(duplicate_keys)})", style="yellow"))
+
     @db_group.command("render-md")
     @click.option("-i", "--input", "input_path", required=True, help="Input JSON file path")
     @click.option("-d", "--output-dir", "output_dir", default="rendered_md", help="Output directory")

{deepresearch_flow-0.6.0.dist-info → deepresearch_flow-0.6.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deepresearch-flow
-Version: 0.6.0
+Version: 0.6.1
 Summary: Workflow tools for paper extraction, review, and research automation.
 Author-email: DengQi <dengqi935@gmail.com>
 License: MIT License
@@ -400,6 +400,41 @@ uv run deepresearch-flow paper db merge templates \
 
 Note: `paper db merge` is now split into `merge library` and `merge templates`.
 
+### Merge multiple databases (PDF + Markdown + BibTeX)
+
+```bash
+# 1) Copy PDFs into a single folder
+rsync -av ./pdfs_a/ ./pdfs_merged/
+rsync -av ./pdfs_b/ ./pdfs_merged/
+
+# 2) Copy Markdown folders into a single folder
+rsync -av ./md_a/ ./md_merged/
+rsync -av ./md_b/ ./md_merged/
+
+# 3) Merge JSON libraries
+uv run deepresearch-flow paper db merge library \
+  --inputs ./paper_infos_a.json \
+  --inputs ./paper_infos_b.json \
+  --output ./paper_infos_merged.json
+
+# 4) Merge BibTeX files
+uv run deepresearch-flow paper db merge bibtex \
+  -i ./library_a.bib \
+  -i ./library_b.bib \
+  -o ./library_merged.bib
+```
+
+### Merge BibTeX files
+
+```bash
+uv run deepresearch-flow paper db merge bibtex \
+  -i ./library_a.bib \
+  -i ./library_b.bib \
+  -o ./library_merged.bib
+```
+
+Duplicate keys keep the entry with the most fields; ties keep the first input order.
+
 ### Recommended: Merge templates then filter by BibTeX
 
 ```bash

{deepresearch_flow-0.6.0.dist-info → deepresearch_flow-0.6.1.dist-info}/RECORD

@@ -4,7 +4,7 @@ deepresearch_flow/cli.py,sha256=t4oowCNWldL0DrVJ4d0UlRkuGU2qHej_G0mAc_quteQ,455
 deepresearch_flow/paper/__init__.py,sha256=sunaOkcgAJBrfmcaJTumcWbPGVUSGWvOv2a2Yidzy0A,43
 deepresearch_flow/paper/cli.py,sha256=68d-yccScU0yL6d7eqZVdudPO6i_in8F4v-hKDWILMo,13647
 deepresearch_flow/paper/config.py,sha256=V7z4ApPXCV1acSl2FU3nZGq6nt8uisMhm0GtOq5zzmg,12021
-deepresearch_flow/paper/db.py,sha256=UL2q4CFI33a3DZsZ42VOS_3FtTORnQuAogUfzPVjcO0,86579
+deepresearch_flow/paper/db.py,sha256=Bxhrd8NCaPZ9Ijtp1uiOplwh2Uy0n2Dyn1bO0d4A9bE,91780
 deepresearch_flow/paper/db_ops.py,sha256=cb64jn2ax39i3tCS-0DYmlsJdGX3uBS2u5ncUIbUBic,73980
 deepresearch_flow/paper/extract.py,sha256=78ASAyNLfCl1AsAk2o_v1vskZCNZuayaCHgr0S4V2Vs,87632
 deepresearch_flow/paper/llm.py,sha256=mHfs5IkT3Q6BOh46MDlfUmgVTX24WRf0IKKoOnN8nV8,4007
@@ -463,9 +463,9 @@ deepresearch_flow/translator/placeholder.py,sha256=mEgqA-dPdOsIhno0h_hzfpXpY2asb
 deepresearch_flow/translator/prompts.py,sha256=EvfBvBIpQXARDj4m87GAyFXJGL8EJeahj_rOmp9mv68,5556
 deepresearch_flow/translator/protector.py,sha256=yUMuS2FgVofK_MRXrcauLRiwNvdCCjNAnh6CcNd686o,11777
 deepresearch_flow/translator/segment.py,sha256=rBFMCLTrvm2GrPc_hNFymi-8Ih2DAtUQlZHCRE9nLaM,5146
-deepresearch_flow-0.6.0.dist-info/licenses/LICENSE,sha256=hT8F2Py1pe6flxq3Ufdm2UKFk0B8CBm0aAQfsLXfvjw,1063
-deepresearch_flow-0.6.0.dist-info/METADATA,sha256=fyynvn8LYDTZlsIaKDr3SxQbR8nqQSOk3s85ZIh1t6E,25838
-deepresearch_flow-0.6.0.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
-deepresearch_flow-0.6.0.dist-info/entry_points.txt,sha256=1uIKscs0YRMg_mFsg9NjsaTt4CvQqQ_-zGERUKhhL_Y,65
-deepresearch_flow-0.6.0.dist-info/top_level.txt,sha256=qBl4RvPJNJUbL8CFfMNWxY0HpQLx5RlF_ko-z_aKpm0,18
-deepresearch_flow-0.6.0.dist-info/RECORD,,
+deepresearch_flow-0.6.1.dist-info/licenses/LICENSE,sha256=hT8F2Py1pe6flxq3Ufdm2UKFk0B8CBm0aAQfsLXfvjw,1063
+deepresearch_flow-0.6.1.dist-info/METADATA,sha256=y_CHy1YJ-3P31W43Q_fd8dEkznj7LKLRrCF6F-sGHaQ,26696
+deepresearch_flow-0.6.1.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+deepresearch_flow-0.6.1.dist-info/entry_points.txt,sha256=1uIKscs0YRMg_mFsg9NjsaTt4CvQqQ_-zGERUKhhL_Y,65
+deepresearch_flow-0.6.1.dist-info/top_level.txt,sha256=qBl4RvPJNJUbL8CFfMNWxY0HpQLx5RlF_ko-z_aKpm0,18
+deepresearch_flow-0.6.1.dist-info/RECORD,,