Added a new constant, MAX_TEXT_LENGTH, to the SentenceExtractor class. If the length of the text input exceeds this limit, the text is truncated to fit within the limit. This modification is designed to prevent excessive resource usage for unusually long text inputs.
Added a new constant, MAX_TEXT_LENGTH, to the SentenceExtractor class. If the length of the text input exceeds this limit, the text is truncated to fit within the limit. This modification is designed to prevent excessive resource usage for unusually long text inputs.
There was as bug where if the input of ResultValuator.normalize() was negative, it was truncated to zero. This meant that "bad" results always rank the same. The penalty factor "overallPart" was moved outside of the function and was re-weighted to accomplish a better normalization.
Some of the weights were also re-adjusted based on what appears to produce better results. Needs evaluation.
This seems like it would make the wikipedia search result worse, but it drastically improves the result quality!
This is because wikipedia has a lot of articles that each talk about a lot of irrelevant concepts, and indexing the entire document means tangentially relevant results tend to displace the most relevant results.
There was as bug where if the input of ResultValuator.normalize() was negative, it was truncated to zero. This meant that "bad" results always rank the same. The penalty factor "overallPart" was moved outside of the function and was re-weighted to accomplish a better normalization.
Some of the weights were also re-adjusted based on what appears to produce better results. Needs evaluation.
This seems like it would make the wikipedia search result worse, but it drastically improves the result quality!
This is because wikipedia has a lot of articles that each talk about a lot of irrelevant concepts, and indexing the entire document means tangentially relevant results tend to displace the most relevant results.
Modify processingiterator to be constructed via a factory, to enable re-use of its backing executor service.
This reduces thread churn in the converter sideloader style processing of regular crawl data.
Route the sizeHint from the input parquet file to SideloadProcessing, so that it can set sideloadSizeAdvice appropriately, instead of using a fixed "large" number.
This is necessary to populate the KNOWN_URL column in the domain data table, which is important as it is used in e.g. calculating how far to re-crawl the site in the future.
The URI query string is now URL encoded in the WarcProtocolReconstructor. This change ensures proper encoding of special characters as per the standard URL encoding rules and improves URL validity during the crawling process.
Updated ProcessingIterator's queue polling from one second to 50 milliseconds for improved performance. This facilitates faster document processing across more cores, reducing bottlenecks and slow single-threaded processing.
Use ProcessingIterator to fan out processing of documents across more cores, instead of doing all of it in the writer thread blocking everything else with slow single-threaded processing.
This commit adds a safety check that the URL of the document is from the correct domain.
It also adds a sizeHint() method to SerializableCrawlDataStream which *may* provide an indication if the stream is very large and benefits from sideload-style processing (which is slow).
It furthermore addresses a bug where the ProcessedDomain.write() invoked the wrong method on ConverterBatchWriter and only wrote the domain metadata, not the rest...
The processor normally retains the domain data in memory after processing to be able to do additional site-wide analysis. This works well, except there are a number of outlier websites that have an absurd number of documents that can rapidly fill up the heap of the process.
These websites now receive a simplified treatment. This is executed in the converter batch writer thread. This is slower, but the documents will not be persisted in memory.
With the new crawler modifications, the crawl data comes in a slightly different order, and a result of this is that we can optimize the converter. This is a breaking change that will be incompatible with the old style of crawl data, hence it will linger as a branch for a while.
The first step is to move stuff out of the domain processor into the document processor.
Guava's hashers are a bit allocation hungry, and a big driver of GC churn in the crawler. This switches to the modified Murmur hash function used throughout Marginalia.
Modified site info feed template to secure the description field against injected code. Also adjusted search service by extracting samples within the correct scope and including them in the returned site info. This improves the quality and security of the displayed information.
This change integrates the Feedlot RSS Bot with Marginalia's site info view to offer a preview of the latest updates.
The change introduces a new tiny feature that is a feedlot-client based on Java's HttpClient.
This is for filtering results on how many times the term appears on the domain. The intent is to be beneficial in creating e.g. a domain search feature. It's also very helpful when tracking down spammy domains.
A number of crawl jobs get stuck at about 300 documents, or just under. This seems to be because we fail to increase the crawl limit, which is based on MAX(200, 1.25 x GOOD_URLS) with a 1.5x modifier applied upon a recrawl. GOOD_URLS is based on how many documents successfully process, which is typically fairly small. Switching to KNOWN_URLS should let this grow faster.
The SQL query in the DbCrawlSpecProvider class has been updated; 'GOOD_URLS' has been replaced with 'KNOWN_URLS'. This update ensures the correct data is selected from the DOMAIN_METADATA table.
The floor is also increased to 250 from 200.
Added functionality to remove processes from listing that have not checked in for over a day. A 'removeProcessHeartbeat' function was created to delete the respective entry from the PROCESS_HEARTBEAT table in case heartbeats are absent for more than one day.
This fixes a bug where a prepared statement was created before the table it was supposed to insert into was created. This fails and does nothing.
Furthermore, added the logging that would have warned about this failure, had it been in place.
Since the sideloaders don't populate the documents list in ProcessedDomain to keep the memory footprint manageable, the code that estimates knownUrls etc. will set them to zero, which has negative effects on their ranking. This change will populate them with a bullshit value within a sane ballpark, ensuring that these domains show up in the rankings.
Make some temporary modifications to the CrawledDocument model to support both a "big string" style headers field like in the old formats, and explicit fields as in the new formats. This is a bit awkward to deal with, but it's a necessity until we migrate off the old formats entirely.
The commit also adds a few tests to this logic.
The size of the ArrayBlockingQueue in ConverterWriter.java has been reduced from 4 to 1. This change aims to reduce the memory utilization by not having fully processed domains piling up in RAM. This may cause the writer to go idle in waiting for new data, but that may be preferable to an OOM.
Initialization parameters in DomainLoaderService and DomainIdRegistry have been updated to improve performance. This is done by adding sane default sizes to the hash tables involved, reducing GC churn, but also by setting a sensible fetch size to the queries used, and not fetching irrelevant information such as the domain name.
We do both ip2location and ASN data.
The change also adds some keywords based on autonomous system information, on a somewhat experimental basis. It would be neat to be able to e.g. exclude cloud services or just e.g. cloudflare from the search results.
This variable had a very confusing name, and was dangerously easy to use in the wrong place with the result of getting something that only works as expected half the time.
Ideally this class needs an overhaul, the assumptions it makes about domain names aren't great.
This variable had a very confusing name, and was dangerously easy to use in the wrong place with the result of getting something that only works as expected half the time.
Ideally this class needs an overhaul, the assumptions it makes about domain names aren't great.
This had the knock-on effect of breaking the anchor tag loading in the processor for a lot of domains, since they'd grab domains for the wrong domain name.
In encyclopedia, add a class "mw-content-text" that the WikiSpecialization class is looking for during pruning to give the articles a more fair treatment.
Also add generator keywords based on the generator type provided, to ensure that these documents show up in appropriate filters.
Further, add a new document flag value 'Sideloaded' to be able to distinguish these entries.
This is caused by a resource contention with the query code. The proper way to fix this is to use some form of synchronization, but that will slow the code down. So we just hammer it a few times and let the GC deal with the problem if it fails. Not optimal, but fast.
The code now intercepts and deals with potential exceptions during the parsing of search parameters. This is in response to constant bad requests from bots which were cluttering the logs. A catch clause is added that suppresses these errors and redirects to the base URL.
The site info view can't blindly assume that every website supports https. To figure out which schema to use when linking to a site, execute a single-result search for site:domain.name and then grab the schema off the result.
To allow this, a count parameter is introduced to doSiteSearch() in SearchOperator.
There really is no fantastic place to put this logic, but we need to remove entries with an X-Robots-Tags header where that header indicates it doesn't want to be crawled by Marginalia.
We want to mute some of these records so that they don't produce documents, but in some cases we want a document to be produced for accounting purposes.
Added improved tests that reach for known resources on www.marginalia.nu to test the behavior when encountering bad content type and 404s.
The commit also adds some safety try-catch:es around the charset handling, as it may sometimes explode when fed incorrect data, and we do be guessing...
This commit updates CrawlingThenConvertingIntegrationTest with additional tests for invalid, redirecting, and blocked domains. Improvements have also been made to filter out irrelevant entries in ParquetSerializableCrawlDataStream.
This update includes the addition of timestamps to the parquet format for crawl data, as extracted from the Warc stream.
The parquet format stores the timestamp as a 64 bit long, seconds since unix epoch, without a logical type. This is to avoid having to do format conversions when writing and reading the data.
This parquet field populates the timestamp field in CrawledDocument.
