refactor: Make formatters functions

Formatters have been classes so far which contained some data (the
tamplate to use for formatting and the annotations and documents to
format) and the actual formatting logic (an execute function).

However, we can inject the annotations to be formatted and the templates
so far are static only, so they can be simple variables (we can think
about how to inject them at another point should it come up, no
bikeshedding now).

This way, we can simply pass around one function per formatter, which
should make the code much lighter, easier to add to and especially less
stateful which means less areas of broken interactions to worry about.

papis_extract/__init__.py

@@ -8,12 +8,7 @@ import papis.strings
 from papis.document import Document
 
 from papis_extract import extractor, exporter
-from papis_extract.formatter import (
-    CountFormatter,
-    CsvFormatter,
-    MarkdownFormatter,
-    Formatter,
-)
+from papis_extract.formatter import Formatter, format_count, format_csv, format_markdown
 
 logger = papis.logging.get_logger(__name__)
 
@@ -82,11 +77,11 @@ def main(
         return
 
     if template == "csv":
-        formatter = CsvFormatter()
+        formatter = format_csv
     elif template == "count":
-        formatter = CountFormatter()
+        formatter = format_count
     else:
-        formatter = MarkdownFormatter()
+        formatter = format_markdown
 
     run(documents, edit=manual, write=write, git=git, formatter=formatter)
 
@@ -98,8 +93,10 @@ def run(
     write: bool = False,
     git: bool = False,
 ) -> None:
-    formatter.annotated_docs = extractor.start(documents)
+    annotated_docs = extractor.start(documents)
     if write:
-        exporter.to_notes(formatter, edit=edit, git=git)
+        exporter.to_notes(
+            formatter=formatter, annotated_docs=annotated_docs, edit=edit, git=git
+        )
     else:
-        exporter.to_stdout(formatter)
+        exporter.to_stdout(formatter=formatter, annotated_docs=annotated_docs)

papis_extract/exporter.py

@@ -6,33 +6,35 @@ import papis.api
 import papis.git
 import papis.config
 import Levenshtein
+from papis_extract.annotation import AnnotatedDocument
 
 from papis_extract.formatter import Formatter
 
 logger = papis.logging.get_logger(__name__)
 
 
-def to_stdout(formatter: Formatter) -> None:
+def to_stdout(formatter: Formatter, annotated_docs: list[AnnotatedDocument]) -> 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 = formatter.execute()
-    print(output.rstrip('\n'))
+    output: str = formatter(annotated_docs)
+    print(output.rstrip("\n"))
 
 
-def to_notes(formatter: Formatter, edit: bool, git: bool) -> None:
+def to_notes(
+    formatter: Formatter, annotated_docs: list[AnnotatedDocument], 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 = formatter.annotated_docs
     for entry in annotated_docs:
-        formatted_annotations = formatter.execute(entry).split("\n")
+        formatted_annotations = formatter([entry]).split("\n")
         if formatted_annotations:
             _add_annots_to_note(entry.document, formatted_annotations)
 
@@ -67,7 +69,8 @@ def _add_annots_to_note(
         # add newline if theres no empty space at file end
         if len(existing) > 0 and existing[-1].strip() != "":
             f.write("\n")
-        f.write("\n".join(new_annotations))
+        print(new_annotations)
+        f.write("\n\n".join(new_annotations))
         f.write("\n")
         logger.info(
             f"Wrote {len(new_annotations)} "

papis_extract/formatter.py

@@ -1,114 +1,74 @@
-from dataclasses import dataclass, field
-from typing import Protocol
+from collections.abc import Callable
 
 from papis_extract.annotation import AnnotatedDocument
 
-
-@dataclass
-class Formatter(Protocol):
-    annotated_docs: list[AnnotatedDocument]
-    header: str
-    string: str
-    footer: str
-
-    def execute(self, doc: AnnotatedDocument | None = None) -> str:
-        raise NotImplementedError
+Formatter = Callable[[list[AnnotatedDocument]], str]
 
 
-@dataclass
-class MarkdownFormatter:
-    annotated_docs: list[AnnotatedDocument] = field(default_factory=lambda: list())
-    header: str = ""
-    string: str = (
+def format_markdown(docs: list[AnnotatedDocument] = []) -> str:
+    template = (
         "{{#tag}}#{{tag}}\n{{/tag}}"
-        "{{#quote}}> {{quote}}{{/quote}} {{#page}}[p. {{page}}]{{/page}}\n"
-        "{{#note}}  NOTE: {{note}}{{/note}}"
+        "{{#quote}}> {{quote}}{{/quote}} {{#page}}[p. {{page}}]{{/page}}"
+        "\n{{#note}}  NOTE: {{note}}{{/note}}"
     )
-    footer: str = ""
+    output = ""
+    for entry in docs:
+        if not entry.annotations:
+            continue
 
-    def execute(self, doc: AnnotatedDocument | None = None) -> str:
-        output = ""
-        documents = self.annotated_docs if doc is None else [doc]
-        last = documents[-1]
-        for entry in documents:
-            if not entry.annotations:
-                continue
+        title_decoration = (
+            f"{'=' * len(entry.document.get('title', ''))}   "
+            f"{'-' * len(entry.document.get('author', ''))}"
+        )
+        output += (
+            f"{title_decoration}\n"
+            f"{entry.document['title']} - {entry.document['author']}\n"
+            f"{title_decoration}\n\n"
+        )
+        for a in entry.annotations:
+            output += a.format(template)
+            output += "\n"
 
-            title_decoration = (
-                f"{'=' * len(entry.document.get('title', ''))}   "
-                f"{'-' * len(entry.document.get('author', ''))}"
-            )
-            output += (
-                f"{title_decoration}\n"
-                f"{entry.document['title']} - {entry.document['author']}\n"
-                f"{title_decoration}\n\n"
-            )
-            for a in entry.annotations:
-                output += a.format(self.string)
+        output += "\n\n\n"
 
-            if entry != last:
-                output += "\n\n\n"
-
-        return output
+    return output
 
 
-@dataclass
-class CountFormatter:
-    annotated_docs: list[AnnotatedDocument] = field(default_factory=lambda: list())
-    header: str = ""
-    string: str = ""
-    footer: str = ""
+def format_count(docs: list[AnnotatedDocument] = []) -> str:
+    output = ""
+    for entry in docs:
+        if not entry.annotations:
+            continue
 
-    def execute(self, doc: AnnotatedDocument | None = None) -> str:
-        documents = self.annotated_docs if doc is None else [doc]
-        output = ""
-        for entry in documents:
-            if not entry.annotations:
-                continue
+        count = 0
+        for _ in entry.annotations:
+            count += 1
 
-            count = 0
-            for _ in entry.annotations:
-                count += 1
+        d = entry.document
+        output += (
+            f"{d['author'] if 'author' in d else ''}"
+            f"{' - ' if 'author' in d else ''}"  # only put separator if author
+            f"{entry.document['title'] if 'title' in d else ''}: "
+            f"{count}\n"
+        )
 
-            d = entry.document
-            output += (
-                f"{d['author'] if 'author' in d else ''}"
-                f"{' - ' if 'author' in d else ''}"  # only put separator if author
-                f"{entry.document['title'] if 'title' in d else ''}: "
-                f"{count}\n"
-            )
-
-        return output
+    return output
 
 
-@dataclass
-class CsvFormatter:
-    annotated_docs: list[AnnotatedDocument] = field(default_factory=lambda: list())
+def format_csv(docs: list[AnnotatedDocument] = []) -> str:
     header: str = "type,tag,page,quote,note,author,title,ref,file"
-    string: str = (
+    template: str = (
         '{{type}},{{tag}},{{page}},"{{quote}}","{{note}}",'
         '"{{doc.author}}","{{doc.title}}","{{doc.ref}}","{{file}}"'
     )
-    footer: str = ""
+    output = f"{header}\n"
+    for entry in docs:
+        if not entry.annotations:
+            continue
 
-    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(template, doc=d)
+            output += "\n"
 
-            d = entry.document
-            for a in entry.annotations:
-                output += a.format(self.string, doc=d)
-                output += "\n"
-
-        return output
-
-
-@dataclass
-class CustomFormatter:
-    def __init__(self, header: str = "", string: str = "", footer: str = "") -> None:
-        self.header = header
-        self.string = string
-        self.footer = footer
+    return output