Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update command palette search to prioritize "longest substring" match. #18700

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
+903 −121
Open

Update command palette search to prioritize "longest substring" match. #18700

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions .github/actions/spelling/excludes.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -133,3 +133,4 @@ Resources/(?!en)
^\Qsrc/terminal/parser/ft_fuzzwrapper/run.bat\E$
^\Qsrc/tools/lnkd/lnkd.bat\E$
^\Qsrc/tools/pixels/pixels.bat\E$
^\Qsrc/cascadia/ut_app/FzfTests.cpp\E$
4 changes: 4 additions & 0 deletions .github/actions/spelling/expect/expect.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -649,6 +649,7 @@ FONTSTRING
FONTTYPE
FONTWIDTH
FONTWINDOW
foob
FORCEOFFFEEDBACK
FORCEONFEEDBACK
FRAMECHANGED
Expand All @@ -666,9 +667,11 @@ fuzzer
fuzzmain
fuzzmap
fuzzwrapper
fuzzyfinder
fwdecl
fwe
fwlink
fzf
gci
gcx
gdi
Expand Down Expand Up @@ -1246,6 +1249,7 @@ onecoreuuid
ONECOREWINDOWS
onehalf
oneseq
oob
openbash
opencode
opencon
Expand Down
196 changes: 75 additions & 121 deletions src/cascadia/TerminalApp/FilteredCommand.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@
#include "CommandPalette.h"
#include "HighlightedText.h"
#include <LibraryResources.h>
#include "fzf/fzf.h"

#include "FilteredCommand.g.cpp"

Expand Down Expand Up @@ -57,15 +58,16 @@ namespace winrt::TerminalApp::implementation
if (filter != _Filter)
{
Filter(filter);
HighlightedName(_computeHighlightedName());
Weight(_computeWeight());
HighlightedName(_computeHighlightedName());
}
}

// Method Description:
// - Looks up the filter characters within the item name.
// Iterating through the filter and the item name it tries to associate the next filter character
// with the first appearance of this character in the item name suffix.
// Using the fzf algorithm to traceback from the maximum score to highlight the chars with the
// optimal match. (Preference is given to word boundaries, consecutive chars and special characters
// while penalties are given for gaps)
//
// E.g., for filter="c l t s" and name="close all tabs after this", the match will be "CLose TabS after this".
//
Expand All @@ -77,85 +79,45 @@ namespace winrt::TerminalApp::implementation
//
// E.g., ("CL", true) ("ose ", false), ("T", true), ("ab", false), ("S", true), ("after this", false)
//
// TODO: we probably need to merge this logic with _getWeight computation?
//
// Return Value:
// - The HighlightedText object initialized with the segments computed according to the algorithm above.
winrt::TerminalApp::HighlightedText FilteredCommand::_computeHighlightedName()
{
const auto segments = winrt::single_threaded_observable_vector<winrt::TerminalApp::HighlightedTextSegment>();
auto commandName = _Item.Name();
auto isProcessingMatchedSegment = false;
uint32_t nextOffsetToReport = 0;
uint32_t currentOffset = 0;

for (const auto searchChar : _Filter)
if (Weight() == 0)
{
const WCHAR searchCharAsString[] = { searchChar, L'\0' };
while (true)
{
if (currentOffset == commandName.size())
{
// There are still unmatched filter characters but we finished scanning the name.
// In this case we return the entire item name as unmatched
auto entireNameSegment{ winrt::make<HighlightedTextSegment>(commandName, false) };
segments.Clear();
segments.Append(entireNameSegment);
return winrt::make<HighlightedText>(segments);
}

// GH#9941: search should be locale-aware as well
// We use the same comparison method as upon sorting to guarantee consistent behavior
const WCHAR currentCharAsString[] = { commandName[currentOffset], L'\0' };
auto isCurrentCharMatched = lstrcmpi(searchCharAsString, currentCharAsString) == 0;
if (isProcessingMatchedSegment != isCurrentCharMatched)
{
// We reached the end of the region (matched character came after a series of unmatched or vice versa).
// Conclude the segment and add it to the list.
// Skip segment if it is empty (might happen when the first character of the name is matched)
auto sizeToReport = currentOffset - nextOffsetToReport;
if (sizeToReport > 0)
{
winrt::hstring segment{ commandName.data() + nextOffsetToReport, sizeToReport };
auto highlightedSegment{ winrt::make<HighlightedTextSegment>(segment, isProcessingMatchedSegment) };
segments.Append(highlightedSegment);
nextOffsetToReport = currentOffset;
}
isProcessingMatchedSegment = isCurrentCharMatched;
}

currentOffset++;

if (isCurrentCharMatched)
{
// We have matched this filter character, let's move to matching the next filter char
break;
}
}
segments.Append(winrt::TerminalApp::HighlightedTextSegment(commandName, false));
return winrt::make<HighlightedText>(segments);
}

// Either the filter or the item name were fully processed.
// If we were in the middle of the matched segment - add it.
if (isProcessingMatchedSegment)
auto pattern = fzf::matcher::ParsePattern(Filter());
auto positions = fzf::matcher::GetPositions(commandName, pattern);
// positions are returned is sorted pairs by search term. E.g. sp anta {5,4,11,10,9,8}
// sorting these in ascending order so it is easier to build the text segments
std::ranges::sort(positions, std::less<>());
// a position can be matched in multiple terms, removed duplicates to simplify segments
positions.erase(std::unique(positions.begin(), positions.end()), positions.end());
size_t lastPosition = 0;
for (auto position : positions)
{
auto sizeToReport = currentOffset - nextOffsetToReport;
if (sizeToReport > 0)
if (position > lastPosition)
{
winrt::hstring segment{ commandName.data() + nextOffsetToReport, sizeToReport };
auto highlightedSegment{ winrt::make<HighlightedTextSegment>(segment, true) };
segments.Append(highlightedSegment);
nextOffsetToReport = currentOffset;
hstring nonMatchSegment{ commandName.data() + lastPosition, static_cast<unsigned>(position - lastPosition) };
segments.Append(winrt::TerminalApp::HighlightedTextSegment(nonMatchSegment, false));
}

hstring matchSegment{ commandName.data() + position, 1 };
segments.Append(winrt::TerminalApp::HighlightedTextSegment(matchSegment, true));

lastPosition = position + 1;
}

// Now create a segment for all remaining characters.
// We will have remaining characters as long as the filter is shorter than the item name.
auto sizeToReport = commandName.size() - nextOffsetToReport;
if (sizeToReport > 0)
if (lastPosition < commandName.size())
{
winrt::hstring segment{ commandName.data() + nextOffsetToReport, sizeToReport };
auto highlightedSegment{ winrt::make<HighlightedTextSegment>(segment, false) };
segments.Append(highlightedSegment);
hstring segment{ commandName.data() + lastPosition, static_cast<unsigned>(commandName.size() - lastPosition) };
segments.Append(winrt::TerminalApp::HighlightedTextSegment(segment, false));
}

return winrt::make<HighlightedText>(segments);
Expand All @@ -164,68 +126,60 @@ namespace winrt::TerminalApp::implementation
// Function Description:
// - Calculates a "weighting" by which should be used to order a item
// name relative to other names, given a specific search string.
// Currently, this is based off of two factors:
// * The weight is incremented once for each matched character of the
// search text.
// * If a matching character from the search text was found at the start
// of a word in the name, then we increment the weight again.
// * For example, for a search string "sp", we want "Split Pane" to
// appear in the list before "Close Pane"
// * Consecutive matches will be weighted higher than matches with
// characters in between the search characters.
// - This will return 0 if the item should not be shown. If all the
// characters of search text appear in order in `name`, then this function
// will return a positive number. There can be any number of characters
// separating consecutive characters in searchText.
// * For example:
// "name": "New Tab"
// "name": "Close Tab"
// "name": "Close Pane"
// "name": "[-] Split Horizontal"
// "name": "[ | ] Split Vertical"
// "name": "Next Tab"
// "name": "Prev Tab"
// "name": "Open Settings"
// "name": "Open Media Controls"
// * "open" should return both "**Open** Settings" and "**Open** Media Controls".
// * "Tab" would return "New **Tab**", "Close **Tab**", "Next **Tab**" and "Prev
// **Tab**".
// * "P" would return "Close **P**ane", "[-] S**p**lit Horizontal", "[ | ]
// S**p**lit Vertical", "**P**rev Tab", "O**p**en Settings" and "O**p**en Media
// Controls".
// * "sv" would return "[ | ] Split Vertical" (by matching the **S** in
// "Split", then the **V** in "Vertical").
// Currently, this uses a derivative of the fzf implementation of the Smith Waterman algorithm:
// - This will return 0 if the item should not be shown.
//
// Factors that affect a score (Taken from the fzf repository)
// Scoring criteria
// ----------------
//
// - We prefer matches at special positions, such as the start of a word, or
// uppercase character in camelCase words.
// - Note everything is converted to lower case so this does not apply
//
// - That is, we prefer an occurrence of the pattern with more characters
// matching at special positions, even if the total match length is longer.
// e.g. "fuzzyfinder" vs. "fuzzy-finder" on "ff"
// ````````````
// - Also, if the first character in the pattern appears at one of the special
// positions, the bonus point for the position is multiplied by a constant
// as it is extremely likely that the first character in the typed pattern
// has more significance than the rest.
// e.g. "fo-bar" vs. "foob-r" on "br"
// ``````
// - But since fzf is still a fuzzy finder, not an acronym finder, we should also
// consider the total length of the matched substring. This is why we have the
// gap penalty. The gap penalty increases as the length of the gap (distance
// between the matching characters) increases, so the effect of the bonus is
// eventually cancelled at some point.
// e.g. "fuzzyfinder" vs. "fuzzy-blurry-finder" on "ff"
// ```````````
// - Consequently, it is crucial to find the right balance between the bonus
// and the gap penalty. The parameters were chosen that the bonus is cancelled
// when the gap size increases beyond 8 characters.
//
// - The bonus mechanism can have the undesirable side effect where consecutive
// matches are ranked lower than the ones with gaps.
// e.g. "foobar" vs. "foo-bar" on "foob"
// ```````
// - To correct this anomaly, we also give extra bonus point to each character
// in a consecutive matching chunk.
// e.g. "foobar" vs. "foo-bar" on "foob"
// ``````
// - The amount of consecutive bonus is primarily determined by the bonus of the
// first character in the chunk.
// e.g. "foobar" vs. "out-of-bound" on "oob"
// ````````````
// Arguments:
// - searchText: the string of text to search for in `name`
// - name: the name to check
// Return Value:
// - the relative weight of this match
int FilteredCommand::_computeWeight()
{
auto result = 0;
auto isNextSegmentWordBeginning = true;

for (const auto& segment : _HighlightedName.Segments())
{
const auto& segmentText = segment.TextSegment();
const auto segmentSize = segmentText.size();

if (segment.IsHighlighted())
{
// Give extra point for each consecutive match
result += (segmentSize <= 1) ? segmentSize : 1 + 2 * (segmentSize - 1);

// Give extra point if this segment is at the beginning of a word
if (isNextSegmentWordBeginning)
{
result++;
}
}

isNextSegmentWordBeginning = segmentSize > 0 && segmentText[segmentSize - 1] == L' ';
}

return result;
auto pattern = fzf::matcher::ParsePattern(Filter());
auto score = fzf::matcher::GetScore(Item().Name(), pattern);
return score;
}

// Function Description:
Expand Down
3 changes: 3 additions & 0 deletions src/cascadia/TerminalApp/TerminalAppLib.vcxproj
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -138,6 +138,8 @@
</ClInclude>
<ClInclude Include="FilteredCommand.h" />
<ClInclude Include="Pane.h" />
<ClInclude Include="fzf/fzf.h" />
<ClInclude Include="fzf/LICENSE" />
<ClInclude Include="ColorHelper.h" />
<ClInclude Include="pch.h" />
<ClInclude Include="ShortcutActionDispatch.h">
Expand Down Expand Up @@ -212,6 +214,7 @@
<ClCompile Include="TabBase.cpp">
<DependentUpon>TabBase.idl</DependentUpon>
</ClCompile>
<ClCompile Include="fzf/fzf.cpp" />
<ClCompile Include="TabPaletteItem.cpp" />
<ClCompile Include="TaskbarState.cpp">
<DependentUpon>TaskbarState.idl</DependentUpon>
Expand Down
12 changes: 12 additions & 0 deletions src/cascadia/TerminalApp/TerminalAppLib.vcxproj.filters
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,6 +41,9 @@
<ClCompile Include="HighlightedText.cpp">
<Filter>highlightedText</Filter>
</ClCompile>
<ClCompile Include="fzf/fzf.cpp">
<Filter>fzf</Filter>
</ClCompile>
<ClCompile Include="Toast.cpp" />
<ClCompile Include="LanguageProfileNotifier.cpp" />
<ClCompile Include="Monarch.cpp" />
Expand Down Expand Up @@ -77,6 +80,12 @@
<ClInclude Include="HighlightedText.h">
<Filter>highlightedText</Filter>
</ClInclude>
<ClInclude Include="fzf/fzf.h">
<Filter>fzf</Filter>
</ClInclude>
<ClInclude Include="fzf/LICENSE">
<Filter>fzf</Filter>
</ClInclude>
<ClInclude Include="Toast.h" />
<ClInclude Include="LanguageProfileNotifier.h" />
<ClInclude Include="WindowsPackageManagerFactory.h" />
Expand Down Expand Up @@ -176,6 +185,9 @@
<Filter Include="highlightedText">
<UniqueIdentifier>{e490f626-547d-4b5b-b22d-c6d33c9e3210}</UniqueIdentifier>
</Filter>
<Filter Include="fzf">
<UniqueIdentifier>{e4588ff4-c80a-40f7-be57-3e81f570a93d}</UniqueIdentifier>
</Filter>
</ItemGroup>
<ItemGroup>
<ApplicationDefinition Include="App.xaml">
Expand Down
22 changes: 22 additions & 0 deletions src/cascadia/TerminalApp/fzf/LICENSE
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,22 @@
The MIT License (MIT)

Copyright (c) 2013-2024 Junegunn Choi
Copyright (c) 2021-2025 Simon Hauser

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
Loading
Loading