Sample

Buse/100

Human rating uses the original dataset scale.

DatasetBuse
Human2.13
Lines8

Scores

MethodScoreScaleDataset rank
(0–100)
Posnett00–1; higher is more readable
0.0Δ 4.0
Scalabrino0.2530–1; higher is more readable
1.5Δ 2.5
Dorn0.0970–1; higher is more readable
0.0Δ 4.0
Mi0.0070–1; higher is more readable
0.0Δ 4.0
LLOC7Logical lines; fewer is more readable
4.0Δ 0.0
DeepSeek V4 Pro15.330–20; mean of three runs
65.7Δ 61.6
GPT-6.1 Sol13.330–20; mean of three runs
20.7Δ 16.7
BTTF0.1570–1; higher is more readable
2.0Δ 2.0
Human2.131–5; original rating
4.0

Views

Download official ZIP ↗

Select the downloaded ZIP or its extracted folder. Files stay in your browser.

Loading feature diagnosis…

Source regions

Feature contributions

Feature contributions

FeatureValueContribution
(log-odds)
Halstead volumeHow much symbolic information the code contains. It combines the total number of operators and values with the number of distinct ones.520.921−17.19
Byte entropyHow varied the bytes in the source text are. Repeated bytes give a lower value; a more even mixture gives a higher value.4.465−6.697
Source linesThe number of non-empty source lines.8+3.2

Intercept 8.87 + feature contributions = -11.818 (log-odds).
Score = sigmoid(-11.818) = 0.

Score0.253

Feature contributions

FeatureValueContribution
(log-odds)
Identifier layout (horizontal)How rapidly the visual pattern of identifier characters changes from left to right. It measures spatial frequency, not the meaning of the names.118−1.333
Comment layout (horizontal)How rapidly the visual pattern of comment characters changes from left to right.0−0.804
Comment layout frequencyHow frequently the layout switches between lines containing only comments or whitespace and lines containing code.9+0.718
Indentation frequencyHow rapidly indentation changes from line to line.1−0.703
Identifier layout (vertical)How rapidly the visual pattern of identifier characters changes from top to bottom.7.765−0.301
Number layout frequencyHow rapidly the number of numeric values per line changes through the code.8+0.237
Operator visual areaThe share of the code’s visual area occupied by operators such as +, =, and *.0.109−0.117

Intercept 0.07 + feature contributions = -2.233 (log-odds).
Score = sigmoid(-2.233) = 0.097.

Score0.007
Score7
Run 1

The code shows good structure with a clear method name and purpose. It uses a separate list for removal to avoid concurrent modification, which is a good practice. However, it suffers from nesting with an if statement inside another if, and the logging check adds complexity. The method name is verbose and could be simplified. The code is reasonably concise but could be more efficient.

Run 2

The code is generally readable with a clear purpose: removing old entries from a map. The structure is logical and action granularity is reasonable, with one primary operation per line. However, there is some nesting due to the conditional logging block inside the loop, which slightly interrupts flow. Variable names are descriptive, and logging intent is clear. The code is concise but could be improved with a helper method for logging or extraction of the loop body to reduce nesting. The synchronized keyword and method name indicate purpose well.

Run 3

The code demonstrates generally good readability. It is logically structured with a clear purpose: purging old entries from a map. The method name and variable names are descriptive. Nesting is shallow (two levels), and action granularity is reasonable. The main weaknesses are the incomplete code snippet (ends abruptly and lacks closing braces), the slightly verbose logging condition combined with key collection, and the potential for extraction of the removal logic into a helper method. The synchronization and threshold logic are clear. Overall, it reads fairly naturally despite the truncation.

Run 1

Code Structure: The visible code logically separates identifying stale entries, logging, and collecting removal keys, but the snippet is incomplete and does not show actual removal. Nesting: A loop and two nested conditionals introduce moderate depth, though each level has a clear role. Clarity of intent: Descriptive method and variable names communicate the cleanup purpose; the cutoff expression thisTurn - 1 lacks an explanation. Code length: The snippet is reasonably concise, although the method name and logging statement are lengthy. Action granularity: Each statement performs one action, making individual operations easy to follow. Reading flow: Initialization followed by iteration, filtering, logging, and collection is natural, but missing closing braces and the remainder of the method prevent a complete reading.

Run 2

Code Structure: The visible portion logically initializes a removal list, scans entries, identifies stale messages, and records their keys, but the snippet is incomplete and does not show actual removal or closing braces. Nesting: Three nested blocks add moderate indentation, although each condition is straightforward. Clarity of intent: Descriptive method and variable names communicate the purpose; the meaning of the thisTurn - 1 cutoff is less explicit. Code length: Mostly concise, though explicit intValue() and verbose generic construction add clutter. Action granularity: Each statement performs one clear action. Reading flow: The sequence is natural, but logging interrupts the main filtering action and the truncated ending prevents evaluating the complete flow.

Run 3

Code Structure: The visible portion logically scans ignored messages and collects expired keys, but the snippet is incomplete, with no closing braces or removal phase. Nesting: A loop and two nested conditionals create moderate indentation. Clarity of intent: The method and variable names clearly communicate cleanup, though the thisTurn - 1 cutoff is unexplained. Code length: The visible code is reasonably concise, but explicit intValue() and verbose generic syntax add clutter. Action granularity: Each statement performs one clear action. Reading flow: Iteration, expiration checking, logging, and key collection follow a natural sequence, but the abrupt ending prevents understanding the complete operation.