CodeFinder

CodeFinder is a very fast desktop search tool for programmers, IT investigators and researchers who work with large collections of plain text files.

It is built for repeated searching in the same file sets, such as source code, logs, notes, exports, copied mail text, malware-related text and other plain text data where normal desktop search is often too broad or too slow.

With Saved Search Areas, CodeFinder can keep file contents in RAM, making repeated searches extremely fast, often well below one second once the area is loaded.

Search results are grouped by file and shown with line numbers and highlighted matches. Double-click a result to open the real file directly at the matching line. From there, you can double-click text to use it as the next search term, making it easy to follow class names, function names, member names, namespaces, IP addresses, email addresses and other text trails.

For programmers, this means you can search for a class, jump to a hit, double-click a function name and continue from there. For IT and malware research, it helps you move quickly from one address, name or value to the next without constantly retyping.

Selected result text can also be copied directly with Ctrl+C or Ctrl+Insert, which is useful when you want to reuse an exact function call, path, address or other piece of text.

Find is the text you want to search for.
In Files is a space separated list of file types such as h hpp cpp c or pas dfm dpr. Do not use * and do not use semicolons. Just type the extensions separated by spaces. If the field is empty, all file types are searched.
In Folder can be either a real folder path for a quick one-off search, or the name of a saved Search Area. Internally CodeFinder treats the field as a real folder if it contains \, / or :. If it does not contain those characters, CodeFinder tries to find a saved Search Area with that name.
SubFolders decides whether subfolders are included in a quick folder search.
Match Case is exactly what it sounds like.
Whole Words is more useful than in many other tools. CodeFinder treats the complete search term as the word, even if it contains punctuation. In many search tools a term such as word->word will fail with Whole Words turned on. In CodeFinder the complete text in the search field is the word to be matched, so a search like that still makes sense here.
Wrap at end belongs to the internal viewer search and decides whether the search continues from the top when the end is reached.

Search areas list

The search areas list is the place where saved search areas are shown and can be accessed. By double-clicking an Item in the list it becomes the active search area and as such determine where the next search is performed.
By right—clicking you get the popup menu shown below.
There are four columns in the list, the two at the left with the colored dots tell two different things. There are four columns in the list, the two at the left with the colored dots tell two different things. The first column shows how the area is configured: gray means No Load, yellow means Load on First Use and green means Load at Startup. The second row shows whether the area is loaded right now: green means loaded and gray means not loaded. The Mem column shows how much RAM an area is using when it is loaded. Search Areas really shows their strength when you keep doing searches on the same sets of files.

Search area list popup menu

The popup menu on the Search Areas list gives the main management commands.
You can add a new Search Area, activate the selected one, edit it, delete it, or release the memory used by a loaded area.
Releasing memory is important when you want the area to reread changed files. Loaded areas keep file contents in RAM for speed, but the files are not held open. That means you can still edit the real files in your editor, but the loaded text snapshot will only change after the area is released and loaded again.

Create or edit search area

This window is where the saved setup is defined.
Search Area Name is the text that appears in the main In Folder field and in the Search Areas list. Choose a name that makes sense in daily use, because this is what you will type or select later.
File Types to Search in works the same way as the main In Files field: extensions separated by spaces, no *, no semicolons, and an empty field means all file types.
Directories to Search is one folder per line.
Directory Excludes and File Excludes use simple wildcard matching. word* means text that starts with word, *word means text that ends with word, and *word* means the text can appear anywhere. The same wildcard style is used in both exclude fields.
Search Subfolders includes subfolders for the listed directories.
Include NO extensions includes files that have no extension at all.
No Load reads files when needed. Load on First Use loads the file contents into RAM the first time the area is searched. Load at Startup does it during program start. This lets you balance startup time against search speed.
When an area has been loaded, the text stays ready in RAM for very fast repeated searches. The files are not kept open, so you can still edit them normally. If you change a file that belongs to a loaded area, the changed text will not be searched until the area memory is released and the area is loaded again.
If you edit an existing Search Area and also change its name, the dialog gives you Save As so the changed setup can be stored as a new area instead of overwriting the old one. In practice that also means you must change the name if you want the Save As option to appear.

Small but useful controls

S, M and L switch between small, medium and large program sizes. The chosen size is remembered, so you can keep the view that fits your screen best.

Create Area stores the current search setup as a saved Search Area. It is the quick way to turn a one-off setup into something reusable.

Double-Click Modifiers change the next search term when you double-click text in CodeViewer. This is very useful for things like ., : and ->, where you want to continue the search with a prefix added automatically instead of typing it by hand.

The status line shows search result information such as number of hits, number of files and search time.
When CodeFinder runs in forcewide mode, that also shows in the status line, at the far right side.
CodeFinder has a special forcewide command line mode, for cases where you need the search to be performed in double byte characters, normally CodeFinder handles it's search in ASCII, Utf8 and ANSI codepages.
Example of usage:
CodeFinder.exe forcewide
If you start CodeFinder from a shortcut, add forcewide after the exe path in the shortcut target field.
The program looks the same in forcewide mode as in normal mode. The visible difference is the small forcewide mark in the status line.
If you only need forcewide now and then, the easiest solution is usually to make a second shortcut with that parameter added. You can also run a normal CodeFinder and a forcewide CodeFinder at the same time.

Use saved search areas for anything you search often. That is where CodeFinder becomes much faster than retyping folders and file types every time.
If you are following code structure, start broad and then narrow down by double-clicking new text in CodeViewer. Search for a class, show a hit, double-click a method name, and continue from there.
If you are following indicators in logs or malware related text, use the same idea for names, values, IP addresses and email addresses.
Learn the result shortcuts. Showing the hit, opening your own editor, opening Explorer and copying selected result text from the keyboard saves a surprising amount of time over a normal workday.
If a loaded Search Area seems out of date after you edited files, release the memory for that area and let it load again on the next search.

CodeFinder is meant for plain text files. It is not intended as a binary file searcher.
Case-insensitive searching can run into edge cases with some Unicode letters where upper-case and lower-case forms do not have the same length. That is one of the reasons forcewide exists.
A loaded Search Area is a RAM snapshot for speed. That is by design, but it also means changed files are not automatically reread until the area is released and loaded again.