If you haven’t seen it yet, check out our YouTube video: ‘Good News - FUnLogMatic Just Got Faster.’ There's a link below.
As the video explains, our engineers have achieved a significant breakthrough, reducing the search time of a 1.06GB, 13M line text file from 21 seconds to just 1 second. This is a massive improvement, especially considering that LogMatic is designed to perform multiple searches on a file to extract and process contextual information.
Previously, LogMatic relied on a third-party PCRE2 Xojo library. While faster than Xojo’s built-in regular expression engine, we optimized performance by gathering all searches to be performed directly on the file, invoking them for each line as we read it. This approach reduced file accesses and minimized repeated scans over large files.
You can still use this default implementation by selecting the 'PCRE2' search type radio button.
Thanks to Andrew Gallant for placing ‘ripgrep’ into the public domain via the ‘unlicense,’ we were able to integrate it as a search option. It is incredibly fast.
Ripgrep comes in two flavors, both supported on all FUnLogMatic platforms: macOS M Series, macOS x86_64, Windows x86_64, and Linux x86_64:
RP Search: By selecting the 'RP' search type radio button, the search is performed using ripgrep’s regex syntax. This is the fastest possible search mode, but it doesn’t support some of the advanced PCRE2 capabilities.
RP PCRE2 Search: By selecting the 'RP PCRE2' search type radio button, ripgrep is instructed to use PCRE2 capabilities when needed. This adds some extra overhead to ripgrep’s algorithm but offers greater flexibility.
You can view the regex syntax used by ripgrep here:
We hope you’re as excited as we are about how much this can improve productivity. Take it for a test drive by downloading v1.3 of FUnLogMatic now.