feat: Add CSV formatter

Added formatter for csv-compatible syntax. The formatting is quite basic
with no escaping happening should that be necessary. However, for an
initial csv output it suffices for me.

papis_extract/__init__.py

@@ -8,7 +8,12 @@ import papis.strings
 from papis.document import Document
 
 from papis_extract import extractor, exporter
-from papis_extract.formatter import CountFormatter, MarkdownFormatter, Formatter
+from papis_extract.formatter import (
+    CountFormatter,
+    CsvFormatter,
+    MarkdownFormatter,
+    Formatter,
+)
 
 logger = papis.logging.get_logger(__name__)
 
@@ -80,24 +85,24 @@ def main(
         return
 
     if template == "csv":
-        raise NotImplementedError
+        formatter = CsvFormatter()
     elif template == "count":
         formatter = CountFormatter()
     else:
         formatter = MarkdownFormatter()
 
-    run(documents, edit=manual, write=write, git=git, template=formatter)
+    run(documents, edit=manual, write=write, git=git, formatter=formatter)
 
 
 def run(
     documents: list[Document],
-    template: Formatter,
+    formatter: Formatter,
     edit: bool = False,
     write: bool = False,
     git: bool = False,
 ) -> None:
-    template.annotated_docs = extractor.start(documents)
+    formatter.annotated_docs = extractor.start(documents)
     if write:
-        exporter.to_notes(template, edit=edit, git=git)
+        exporter.to_notes(formatter, edit=edit, git=git)
     else:
-        exporter.to_stdout(template)
+        exporter.to_stdout(formatter)

papis_extract/annotation_data.py

@@ -34,7 +34,7 @@ class Annotation:
     type: str = "Highlight"
     minimum_similarity_color: float = 1.0
 
-    def format(self, template: str, doc: Document = Document()):
+    def format(self, formatting: str, doc: Document = Document()):
         """Return a formatted string of the annotation.
 
         Given a provided formatting pattern, this method returns the annotation
@@ -50,7 +50,7 @@ class Annotation:
             "type": self.type,
             "doc": doc,
         }
-        return chevron.render(template, data)
+        return chevron.render(formatting, data)
 
     @property
     def colorname(self):

papis_extract/exporter.py

@@ -12,27 +12,27 @@ from papis_extract.formatter import Formatter
 logger = papis.logging.get_logger(__name__)
 
 
-def to_stdout(template: Formatter) -> None:
+def to_stdout(formatter: Formatter) -> None:
     """Pretty print annotations to stdout.
 
     Gives a nice human-readable representations of
     the annotations in somewhat of a list form.
     Not intended for machine-readability.
     """
-    output:str = template.execute()
+    output:str = formatter.execute()
     print(output.rstrip('\n'))
 
 
-def to_notes(template: Formatter, edit: bool, git: bool) -> None:
+def to_notes(formatter: Formatter, edit: bool, git: bool) -> None:
     """Write annotations into document notes.
 
     Permanently writes the given annotations into notes
     belonging to papis documents. Creates new notes for
     documents missing a note field or appends to existing.
     """
-    annotated_docs = template.annotated_docs
+    annotated_docs = formatter.annotated_docs
     for entry in annotated_docs:
-        formatted_annotations = template.execute(entry).split("\n")
+        formatted_annotations = formatter.execute(entry).split("\n")
         if formatted_annotations:
             _add_annots_to_note(entry.document, formatted_annotations)
 

papis_extract/formatter.py

@@ -73,7 +73,7 @@ class CountFormatter:
             d = entry.document
             output += (
                 f"{d['author'] if 'author' in d else ''}"
-                f"{' - ' if 'author' in d else ''}" # only put separator if author
+                f"{' - ' if 'author' in d else ''}"  # only put separator if author
                 f"{entry.document['title'] if 'title' in d else ''}: "
                 f"{count}\n"
             )
@@ -83,10 +83,28 @@ class CountFormatter:
 
 @dataclass
 class CsvFormatter:
-    header: str = "type, tag, page, quote, note, file"
+    annotated_docs: list[AnnotatedDocument] = field(default_factory=lambda: list())
-    string: str = "{{type}}, {{tag}}, {{page}}, {{quote}}, {{note}}, {{file}}"
+    header: str = "type,tag,page,quote,note,author,title,ref,file"
+    string: str = (
+        '{{type}},{{tag}},{{page}},"{{quote}}","{{note}}",'
+        '"{{doc.author}}","{{doc.title}}","{{doc.ref}}","{{file}}"'
+    )
     footer: str = ""
 
+    def execute(self, doc: AnnotatedDocument | None = None) -> str:
+        documents = self.annotated_docs if doc is None else [doc]
+        output = f"{self.header}\n"
+        for entry in documents:
+            if not entry.annotations:
+                continue
+
+            d = entry.document
+            for a in entry.annotations:
+                output += a.format(self.string, doc=d)
+                output += "\n"
+
+        return output
+
 
 @dataclass
 class CustomFormatter: