Add MimeType based parser (#4376)

# Add MimeType Based Parser

This PR adds a MimeType Based Parser. The parser inspects the mime-type
of the blob it is parsing and based on the mime-type can delegate to the sub
parser.

## Before submitting

Waiting on adding notebooks until more implementations are landed. 

## Who can review?

Community members can review the PR once tests pass. Tag
maintainers/contributors who might be interested:


@hwchase17
@vowelparrot

tests/unit_tests/document_loader/parsers/test_generic.py

@@ -0,0 +1,95 @@
+"""Module to test generic parsers."""
+
+from typing import Iterator
+
+import pytest
+
+from langchain.document_loaders.base import BaseBlobParser
+from langchain.document_loaders.blob_loaders import Blob
+from langchain.document_loaders.parsers.generic import MimeTypeBasedParser
+from langchain.schema import Document
+
+
+class TestMimeBasedParser:
+    """Test mime based parser."""
+
+    def test_without_fallback_parser(self) -> None:
+        class FirstCharParser(BaseBlobParser):
+            def lazy_parse(self, blob: Blob) -> Iterator[Document]:
+                """Extract the first character of a blob."""
+                yield Document(page_content=blob.as_string()[0])
+
+        class SecondCharParser(BaseBlobParser):
+            def lazy_parse(self, blob: Blob) -> Iterator[Document]:
+                """Extract the second character of a blob."""
+                yield Document(page_content=blob.as_string()[1])
+
+        parser = MimeTypeBasedParser(
+            handlers={
+                "text/plain": FirstCharParser(),
+                "text/html": SecondCharParser(),
+            },
+        )
+
+        blob = Blob(data=b"Hello World", mimetype="text/plain")
+        docs = parser.parse(blob)
+        assert len(docs) == 1
+        doc = docs[0]
+        assert doc.page_content == "H"
+
+        # Check text/html handler.
+        blob = Blob(data=b"Hello World", mimetype="text/html")
+        docs = parser.parse(blob)
+        assert len(docs) == 1
+        doc = docs[0]
+        assert doc.page_content == "e"
+
+        blob = Blob(data=b"Hello World", mimetype="text/csv")
+
+        with pytest.raises(ValueError, match="Unsupported mime type"):
+            # Check that the fallback parser is used when the mimetype is not found.
+            parser.parse(blob)
+
+    def test_with_fallback_parser(self) -> None:
+        class FirstCharParser(BaseBlobParser):
+            def lazy_parse(self, blob: Blob) -> Iterator[Document]:
+                """Extract the first character of a blob."""
+                yield Document(page_content=blob.as_string()[0])
+
+        class SecondCharParser(BaseBlobParser):
+            def lazy_parse(self, blob: Blob) -> Iterator[Document]:
+                """Extract the second character of a blob."""
+                yield Document(page_content=blob.as_string()[1])
+
+        class ThirdCharParser(BaseBlobParser):
+            def lazy_parse(self, blob: Blob) -> Iterator[Document]:
+                """Extract the third character of a blob."""
+                yield Document(page_content=blob.as_string()[2])
+
+        parser = MimeTypeBasedParser(
+            handlers={
+                "text/plain": FirstCharParser(),
+                "text/html": SecondCharParser(),
+            },
+            fallback_parser=ThirdCharParser(),
+        )
+
+        blob = Blob(data=b"Hello World", mimetype="text/plain")
+        docs = parser.parse(blob)
+        assert len(docs) == 1
+        doc = docs[0]
+        assert doc.page_content == "H"
+
+        # Check text/html handler.
+        blob = Blob(data=b"Hello World", mimetype="text/html")
+        docs = parser.parse(blob)
+        assert len(docs) == 1
+        doc = docs[0]
+        assert doc.page_content == "e"
+
+        # Check that the fallback parser is used when the mimetype is not found.
+        blob = Blob(data=b"Hello World", mimetype="text/csv")
+        docs = parser.parse(blob)
+        assert len(docs) == 1
+        doc = docs[0]
+        assert doc.page_content == "l"