feat: ✨ Issue 7 add basic repository browser (#8)

* WIP

* Show units in browser

* Fixed tests

* Add source location

* Show code snippet

* Very basic report browser

.gitignore

@@ -1,6 +1,6 @@
 build/
 dist/
-clim.spec
+main.spec
 codelimit.json
 .coverage
 coverage.xml

build-dist.sh

@@ -0,0 +1,2 @@
+#!/bin/sh
+poetry run pyinstaller --specpath dist -n codelimit -F main.py

codelimit/common/Codebase.py

@@ -6,7 +6,7 @@
 class Codebase:
     def __init__(self):
         self.tree = {'./': SourceFolder()}
-        self.measurements = {}
+        self.measurements: dict[str, list[SourceMeasurement]] = {}
 
     def add_file(self, path: str, measurements: list[SourceMeasurement]):
         self.measurements[path] = measurements
@@ -52,8 +52,8 @@ def all_measurements(self) -> list[SourceMeasurement]:
             result.extend(m)
         return result
 
-    def all_measurements_sorted_by_length(self):
-        return sorted(self.all_measurements(), key=lambda m: m.value)
+    def all_measurements_sorted_by_length_asc(self):
+        return sorted(self.all_measurements(), key=lambda m: m.value, reverse=True)
 
     def total_loc(self) -> int:
         result = 0

codelimit/common/Scanner.py

@@ -7,6 +7,7 @@
 from codelimit.common.Codebase import Codebase
 from codelimit.common.SourceMeasurement import SourceMeasurement
 from codelimit.common.scope_utils import build_scopes
+from codelimit.common.source_utils import get_location_range
 from codelimit.languages.c.CLanguage import CLanguage
 from codelimit.languages.python.PythonLaguage import PythonLanguage
 
@@ -52,5 +53,9 @@ def _scan_file(self, language, root, folder, file):
             measurements = []
             for scope in scopes:
                 length = len(scope)
-                measurements.append(SourceMeasurement(scope.header.tokens[0].location.line, length))
+                unit_name = get_location_range(code, scope.header.tokens[0].location, scope.block.tokens[0].location)
+                unit_name = unit_name.strip().replace('\t', ' ').replace('\n', ' ')
+                start_location = scope.header.tokens[0].location
+                end_location = scope.block.tokens[-1].location
+                measurements.append(SourceMeasurement(unit_name, start_location, end_location, length))
             self.codebase.add_file(rel_path, measurements)

codelimit/common/SourceLocation.py

@@ -9,6 +9,9 @@ def __init__(self, line: int, column: int):
     def __str__(self):
         return f'{{line: {self.line}, column: {self.column}}}'
 
+    def __repr__(self):
+        return self.__str__()
+
     def lt(self, other: SourceLocation):
         return self.line < other.line or (self.line == other.line and self.column < other.column)
 

codelimit/common/SourceMeasurement.py

@@ -1,7 +1,11 @@
 from dataclasses import dataclass
 
+from codelimit.common.SourceLocation import SourceLocation
+
 
 @dataclass
 class SourceMeasurement:
-    start_line: int
+    unit_name: str
+    start: SourceLocation
+    end: SourceLocation
     value: int

codelimit/common/report/Browser.py

@@ -0,0 +1,26 @@
+import os.path
+from pathlib import Path
+
+from InquirerPy import inquirer
+from InquirerPy.base import Choice
+
+from codelimit.common.report.Report import Report
+from codelimit.common.report.ReportUnit import format_report_unit, ReportUnit
+from codelimit.common.source_utils import get_location_range
+
+
+class Browser:
+    def __init__(self, report: Report, path: Path = '.'):
+        self.report = report
+        self.path = path
+
+    def show(self):
+        units = [Choice(value=unit, name=format_report_unit(unit)) for unit in
+                 self.report.all_report_units_sorted_by_length_asc()]
+        while True:
+            selected_unit: ReportUnit = inquirer.select(message='Select unit', choices=units).execute()
+            file_path = os.path.join(self.path, selected_unit['file'])
+            with open(file_path) as file:
+                code = file.read()
+            snippet = get_location_range(code, selected_unit['measurement']['start'], selected_unit['measurement']['end'])
+            print(snippet)

codelimit/common/Report.py → codelimit/common/report/Report.py

@@ -4,6 +4,7 @@
 import plotext
 
 from codelimit.common.Codebase import Codebase
+from codelimit.common.report.ReportUnit import ReportUnit
 from codelimit.common.utils import make_profile
 
 
@@ -18,7 +19,7 @@ def get_average(self):
         return ceil(self.codebase.total_loc() / len(self.codebase.all_measurements()))
 
     def ninetieth_percentile(self):
-        sorted_measurements = self.codebase.all_measurements_sorted_by_length()
+        sorted_measurements = self.codebase.all_measurements_sorted_by_length_asc()
         lines_of_code_90_percent = floor(self.codebase.total_loc() * 0.9)
         smallest_units_loc = 0
         for index, m in enumerate(sorted_measurements):
@@ -27,6 +28,14 @@ def ninetieth_percentile(self):
                 return sorted_measurements[index].value
         return 0
 
+    def all_report_units_sorted_by_length_asc(self) -> list[ReportUnit]:
+        result = []
+        for file, measurements in self.codebase.measurements.items():
+            for m in measurements:
+                result.append(ReportUnit(file, m))
+        result = sorted(result, key=lambda unit: unit.measurement.value, reverse=True)
+        return result
+
     def risk_categories(self):
         return make_profile(self.codebase.all_measurements())
 
@@ -35,4 +44,4 @@ def display_risk_category_plot(self):
         volume = make_profile(self.codebase.all_measurements())
         plotext.title("Most Favored Pizzas in the World")
         plotext.simple_bar(labels, volume, color=[34, 226, 214, 196])
-        plotext.show()
+        plotext.show()

codelimit/common/report/ReportReader.py

@@ -0,0 +1,24 @@
+from json import loads
+
+from codelimit.common.Codebase import Codebase
+from codelimit.common.SourceLocation import SourceLocation
+from codelimit.common.report.Report import Report
+from codelimit.common.SourceMeasurement import SourceMeasurement
+
+
+class ReportReader:
+
+    @staticmethod
+    def from_json(json: str) -> Report:
+        d = loads(json)
+        codebase = Codebase()
+        report = Report(codebase)
+        report.uuid = d['uuid']
+        for k, v in d['codebase']['measurements'].items():
+            measurements: list[SourceMeasurement] = []
+            for m in v:
+                start_location = SourceLocation(m['start']['line'], m['start']['column'])
+                end_location = SourceLocation(m['end']['line'], m['end']['column'])
+                measurements.append(SourceMeasurement(m['unit_name'], start_location, end_location, m['value']))
+            codebase.add_file(k, measurements)
+        return report

codelimit/common/report/ReportUnit.py

@@ -0,0 +1,16 @@
+from dataclasses import dataclass
+
+from codelimit.common.SourceMeasurement import SourceMeasurement
+
+
+@dataclass
+class ReportUnit:
+    file: str
+    measurement: SourceMeasurement
+
+
+def format_report_unit(unit: ReportUnit) -> str:
+    name = unit.measurement.unit_name
+    length = unit.measurement.value
+    prefix = f'[{length:3}]' if length < 61 else '[60+]'
+    return f'{prefix} {name}'

codelimit/common/ReportSerializer.py → codelimit/common/report/ReportWriter.py

@@ -1,10 +1,10 @@
-from codelimit.common.Report import Report
 from codelimit.common.SourceFolder import SourceFolder
 from codelimit.common.SourceFolderEntry import SourceFolderEntry
 from codelimit.common.SourceMeasurement import SourceMeasurement
+from codelimit.common.report.Report import Report
 
 
-class ReportSerializer:
+class ReportWriter:
     def __init__(self, report: Report, pretty_print=True):
         self.report = report
         self.tree = report.codebase.tree
@@ -86,13 +86,17 @@ def _measurement_item_to_json(self, name: str, measurements: list[SourceMeasurem
         json += self._close(']')
         return json
 
+    def _measurement_to_json(self, measurement: SourceMeasurement) -> str:
+        json = ''
+        json += f'{{"unit_name": "{measurement.unit_name}", '
+        json += f'"start": {{"line": {measurement.start.line}, "column": {measurement.start.column}}}, '
+        json += f'"end": {{"line": {measurement.end.line}, "column": {measurement.end.column}}}, '
+        json += f'"value": {measurement.value}}}'
+        return self._line(json)
+
     def _source_folder_entry_to_json(self, entry: SourceFolderEntry) -> str:
         json = f'{{"name": "{entry.name}"'
         if entry.profile:
             json += f', "profile": {entry.profile}'
         json += '}'
         return self._line(json)
-
-    def _measurement_to_json(self, measurement: SourceMeasurement) -> str:
-        json = f'{{"start_line": {measurement.start_line}, "value": {measurement.value}}}'
-        return self._line(json)

codelimit/common/scope_utils.py

@@ -1,6 +1,8 @@
+from codelimit.common.Language import Language
 from codelimit.common.Scope import Scope
 from codelimit.common.TokenRange import TokenRange
-from codelimit.common.Language import Language
+from codelimit.common.token_utils import sort_tokens
+from codelimit.common.utils import delete_indices
 
 
 def build_scopes(language: Language, code: str) -> list[Scope]:
@@ -12,20 +14,24 @@ def build_scopes(language: Language, code: str) -> list[Scope]:
 
 
 def _build_scopes_from_headers_and_blocks(headers: list[TokenRange], blocks: list[TokenRange]) -> list[Scope]:
-    result = []
-    reverse_blocks = blocks[::-1]
-    for header in headers[::-1]:
-        scope_blocks = []
-        for index, block in enumerate(reverse_blocks):
-            if block.lt(header) or block.overlaps(header):
-                reverse_blocks = reverse_blocks[index:]
-                break
-            if block.tokens[0].location.column > header.tokens[0].location.column:
-                scope_blocks = block.tokens + scope_blocks
-            else:
-                scope_blocks = []
-        if len(scope_blocks) > 0:
-            scope_block = TokenRange(scope_blocks)
-            result.append(Scope(header, scope_block))
+    result: list[Scope] = []
+    reverse_headers = headers[::-1]
+    for header in reverse_headers:
+        scope_blocks_indices = _find_scope_blocks_indices(header, blocks)
+        scope_tokens = []
+        for i in scope_blocks_indices:
+            scope_tokens.extend(blocks[i].tokens)
+        scope_tokens = sort_tokens(scope_tokens)
+        scope_block = TokenRange(scope_tokens)
+        result.append(Scope(header, scope_block))
+        blocks = delete_indices(blocks, scope_blocks_indices)
     result.reverse()
     return result
+
+
+def _find_scope_blocks_indices(header: TokenRange, blocks: list[TokenRange]) -> list[int]:
+    blocks_after_header = [i for i in range(len(blocks)) if header.lt(blocks[i])]
+    if len(blocks_after_header) > 0:
+        body_blocks = [i for i in blocks_after_header if blocks[blocks_after_header[0]].overlaps(blocks[i])]
+        return body_blocks
+    return []

codelimit/common/source_utils.py

@@ -2,7 +2,6 @@
 
 from codelimit.common.SourceLocation import SourceLocation
 from codelimit.common.Token import Token
-from codelimit.common.TokenRange import TokenRange
 
 
 def get_newline_indices(code: str) -> list[int]:
@@ -38,10 +37,16 @@ def location_to_index(code: str, position: SourceLocation) -> int:
     return result
 
 
-def get_range(code: str, source_range: TokenRange) -> str:
-    start_index = location_to_index(code, source_range.tokens[0].location)
-    end_index = location_to_index(code, source_range.tokens[-1].location)
-    return code[start_index:end_index + len(source_range.tokens[-1].value)]
+def get_token_range(code: str, start: Token, end: Token) -> str:
+    start_index = location_to_index(code, start.location)
+    end_index = location_to_index(code, end.location)
+    return code[start_index:end_index + len(end.value)]
+
+
+def get_location_range(code: str, start: SourceLocation, end: SourceLocation) -> str:
+    start_index = location_to_index(code, start)
+    end_index = location_to_index(code, end)
+    return code[start_index:end_index]
 
 
 def filter_tokens(tokens: list[Token], predicate: Callable[[Token], bool]) -> list[Token]:

codelimit/common/token_utils.py

@@ -16,3 +16,9 @@ def get_balanced_symbol_token_indices(tokens: list[Token], start: str, end: str,
                 if nested or len(block_starts) == 0:
                     result.append((start_index, index))
     return result
+
+
+def sort_tokens(tokens: list[Token]) -> list[Token]:
+    result = sorted(tokens, key=lambda t: t.location.column)
+    result = sorted(result, key=lambda t: t.location.line)
+    return result

codelimit/common/utils.py

@@ -1,5 +1,5 @@
 import os
-from typing import Union
+from typing import Union, Any
 
 from codelimit.common.SourceMeasurement import SourceMeasurement
 
@@ -34,7 +34,7 @@ def path_has_suffix(path: str, suffixes: Union[str, list[str]]):
         return False
 
 
-def get_parent_folder(path: str) -> Union[str, None]:
+def get_parent_folder(path: str) -> str:
     parts = path.split(os.path.sep)
     if len(parts) == 1:
         return '.'
@@ -45,3 +45,7 @@ def get_parent_folder(path: str) -> Union[str, None]:
 def get_basename(path: str) -> str:
     parts = path.split(os.path.sep)
     return parts[-1]
+
+
+def delete_indices(iterable: list, indices: list[int]) -> list[Any]:
+    return [b for i, b in enumerate(iterable) if i not in indices]

codelimit/languages/c/CScopeExtractor.py

@@ -22,4 +22,7 @@ def has_curly_suffix(index):
 
     def extract_blocks(self, tokens: list[Token]) -> list[TokenRange]:
         balanced_tokens = get_balanced_symbol_token_indices(tokens, '{', '}', True)
-        return [TokenRange(tokens[bt[0]:bt[1] + 1]) for bt in balanced_tokens]
+        blocks = [TokenRange(tokens[bt[0]:bt[1] + 1]) for bt in balanced_tokens]
+        sorted_by_line = sorted(blocks, key=lambda tr: tr.tokens[0].location.line)
+        sorted_by_columns = sorted(sorted_by_line, key=lambda tr: tr.tokens[0].location.column)
+        return sorted_by_columns