feat implement ClozeExtractor (#301)

personal_mnemonic_medium/v2/domain/prompt_source/prompt_extractors/cloze_prompt_extractor.py

@@ -0,0 +1,98 @@
+import hashlib
+import logging
+import re
+from collections.abc import Sequence
+
+from ...prompts.cloze_prompt import ClozePrompt, ClozePromptFromDoc
+from ..document_ingesters.document import Document
+from .base_prompt_extractor import BasePromptExtractor
+
+log = logging.getLogger(__name__)
+
+
+class ClozePromptExtractor(BasePromptExtractor):
+    @staticmethod
+    def _break_string_by_two_or_more_newlines(
+        string: str
+    ) -> list[str]:
+        """Break string into a list by 2+ newlines in a row."""
+        return re.split(r"(\n\n)+", string)
+
+    @staticmethod
+    def _has_cloze(string: str) -> bool:
+        if (
+            len(re.findall(r"{.*}", string)) > 0
+            and "BearID" not in string  # Exclude BearID
+            and "$$" not in string  # Exclude math
+            and r"```" not in string  # Exclude code
+            and "Q." not in string  # Exclude Q&A
+            and "A." not in string  # Exclude Q&A
+        ):
+            return True
+        return False
+
+    @staticmethod
+    def _replace_cloze_id_with_unique(
+        string: str, selected_cloze: str | None = None
+    ) -> str:
+        """Each cloze deletion in a note is numbered sequentially.
+
+        This function ensures that the numbering is based on the content of the cloze deletion, essentially ensuring that if you modify the contents of a cloze, only the scheduling of that specific cloze is changed.
+
+        Args:
+            string (str): The string to replace the cloze id with a unique id.
+            selected_cloze (str, optional): If you only want to replace a specific cloze, pass it here. Defaults to None.
+        """
+        if selected_cloze is not None:
+            selected_clozes = [selected_cloze]
+        else:
+            selected_clozes = re.findall(
+                r"{(?!BearID).[^}]*}", string
+            )
+
+        for cloze in selected_clozes:
+            output_hash = (
+                int(
+                    hashlib.sha256(cloze.encode("utf-8")).hexdigest(),
+                    16,
+                )
+                % 10**3
+            )
+
+            new_cloze = f"{{{{c{output_hash}::{cloze[1:-1]}}}}}"
+
+            string = string.replace(cloze, new_cloze)
+
+        return string
+
+    def extract_prompts(
+        self, document: Document
+    ) -> Sequence[ClozePrompt]:
+        prompts: list[ClozePromptFromDoc] = []
+
+        blocks = self._break_string_by_two_or_more_newlines(
+            document.content
+        )
+
+        for block_string in blocks:
+            if self._has_cloze(block_string):
+                clozes = re.findall(
+                    r"{(?!BearID).[^}]*}", block_string
+                )
+
+                # TODO: https://github.com/MartinBernstorff/personal-mnemonic-medium/issues/300 refactor: move clozeid replacement to AnkiCloze generator class
+                for selected_cloze in clozes:
+                    prompt_content = (
+                        self._replace_cloze_id_with_unique(
+                            block_string,
+                            selected_cloze=selected_cloze,
+                        )
+                    )
+
+                    prompts.append(
+                        ClozePromptFromDoc(
+                            text=prompt_content, source_doc=document
+                        )
+                    )
+
+        return prompts

personal_mnemonic_medium/v2/domain/prompt_source/prompt_extractors/test_cloze_prompt_extractor.py

@@ -0,0 +1,29 @@
+from pathlib import Path
+
+from ..document_ingesters.document import Document
+from .cloze_prompt_extractor import ClozePromptExtractor
+
+
+def test_cloze_prompt_extractor(tmpdir: Path):
+    doc = Document(
+        content=r"""
+
+What is the meaning of life? {42}
+
+
+This is another block without any cloze prompts.
+
+#anki/tag/test_tag
+
+""",
+        source_path=tmpdir / "test.md",
+    )
+
+    extractor = ClozePromptExtractor().extract_prompts(doc)
+
+    assert len(extractor) == 1
+    assert (
+        extractor[0].text
+        == r"What is the meaning of life? {{c734::42}}"
+    )
+    assert extractor[0].tags == ["#anki/tag/test_tag"]

personal_mnemonic_medium/v2/domain/prompts/cloze_prompt.py

@@ -1,13 +1,12 @@
 from collections.abc import Sequence
 from dataclasses import dataclass
 
-from personal_mnemonic_medium.v2.domain.prompts.doc_mixin import (
-    DocMixin,
-)
-
 from ..int_hash_str import int_hash_str
+from ..prompt_source.document_ingesters.document import Document
 from .base_prompt import BasePrompt
 
+# TODO: https://github.com/MartinBernstorff/personal-mnemonic-medium/issues/299 refactor: clean up doc type inheritance
+
 
 @dataclass(frozen=True)
 class ClozePrompt(BasePrompt):
@@ -31,5 +30,15 @@ def tags(self) -> Sequence[str]:
         return self.add_tags
 
 
-class ClozePromptFromDoc(ClozePrompt, DocMixin):
-    ...
+@dataclass(frozen=True)
+class ClozePromptFromDoc(ClozePrompt):
+    text: str
+    source_doc: Document
+
+    @property
+    def uid(self) -> int:
+        return int_hash_str(self.text)
+
+    @property
+    def tags(self) -> Sequence[str]:
+        return self.source_doc.tags