An open source, non-profit web search engine
Use a more specific exception in case we're discarding ones we shouldn't
Daoud Clarkea471bc2437de
diff --git a/mwmbl/indexer/update_pages.py b/mwmbl/indexer/update_pages.pyindex a11a216..d3e7afd 100644--- a/mwmbl/indexer/update_pages.py+++ b/mwmbl/indexer/update_pages.py@@ -6,7 +6,7 @@ from time import sleepfrom mwmbl.database import Databasefrom mwmbl.indexer.indexdb import IndexDatabase-from mwmbl.tinysearchengine.indexer import TinyIndex, Document+from mwmbl.tinysearchengine.indexer import TinyIndex, Document, PageErrordef run_update(index_path):@@ -27,7 +27,7 @@ def run_update(index_path):try:indexer.add_to_page(i, documents)break- except ValueError:+ except PageError:documents = documents[:len(documents)//2]if len(documents) == 0:print("No more space")diff --git a/mwmbl/tinysearchengine/indexer.py b/mwmbl/tinysearchengine/indexer.pyindex 4c9bfef..a9aa9f4 100644--- a/mwmbl/tinysearchengine/indexer.py+++ b/mwmbl/tinysearchengine/indexer.py@@ -36,6 +36,10 @@ class TokenizedDocument(Document):T = TypeVar('T')+class PageError(Exception):+ pass++@dataclassclass TinyIndexMetadata:version: int@@ -68,7 +72,7 @@ def _get_page_data(compressor, page_size, data):def _pad_to_page_size(data: bytes, page_size: int):page_length = len(data)if page_length > page_size:- raise ValueError(f"Data is too big ({page_length}) for page size ({page_size})")+ raise PageError(f"Data is too big ({page_length}) for page size ({page_size})")padding = b'\x00' * (page_size - page_length)page_data = data + paddingreturn page_data@@ -142,7 +146,7 @@ class TinyIndex(Generic[T]):page_index = self.get_key_page_index(key)try:self.add_to_page(page_index, [value])- except ValueError:+ except PageError:passdef add_to_page(self, page_index: int, values: list[T]):