fix: emit unrounded percentage in Overrepresented sequences (Java compat)

src/modules/overrepresented_seqs.rs

@@ -416,17 +416,18 @@ impl QCModule for OverRepresentedSeqs {
                 Some(hit) => hit.to_string(),
                 None => "No Hit".to_string(),
             };
-            // The Java ResultsTable.getValueAt() for percentage does
-            // JAVA COMPAT: Math.round(percentage * 100.0) / 100.0, rounding to 2 decimal places.
-            // The text report then calls String.valueOf() on the Double, producing
-            // Java's Double.toString() format.
-            let rounded_pct = (s.percentage * 100.0).round() / 100.0;
+            // JAVA COMPAT: Java's OverRepresentedSeqs.OverrepresentedSeq stores
+            // the raw double percentage without rounding (see OverRepresentedSeqs.java:253),
+            // and AbstractQCModule.writeTable serializes it via String.valueOf(getValueAt(...))
+            // (AbstractQCModule.java:159), which returns Java's Double.toString() of the
+            // unrounded value (e.g. "7.160449112640348"). Pass the raw percentage to the
+            // formatter; do not round to 2 decimals.
             writeln!(
                 writer,
                 "{}\t{}\t{}\t{}",
                 s.seq,
                 s.count,
-                java_format_double(rounded_pct),
+                java_format_double(s.percentage),
                 source
             )?;
         }

tests/data/gen_realistic.py

@@ -0,0 +1,84 @@
+#!/usr/bin/env python3
+"""Deterministically generate a realistic test FASTQ for fastqc-rust equivalence tests.
+
+Produces 1009 reads of 50bp each at uniform Phred 40, with 5 deliberately
+overrepresented sequences at non-round percentages. Background reads are
+pseudo-random (fixed seed) and below the 0.1% overrepresented threshold.
+
+Designed to expose the percentage-precision bug fixed by ewels/FastQC-Rust#2.
+
+Output is written gzipped (~20 KB vs ~120 KB plain) to keep the repo lean.
+The gzip stream uses mtime=0 and the deterministic content above so byte-
+identical regeneration is possible across machines.
+"""
+import gzip
+import random
+import sys
+
+random.seed(20260426)
+
+QUAL = "I" * 50  # Phred 40
+LEN = 50
+TOTAL = 1009
+
+OVERREP = [
+    ("OVERREP_A_HIGH",  73),  # 73/1009 = 7.23488602...%
+    ("OVERREP_B_MID",   37),  # 37/1009 = 3.66699702...%
+    ("OVERREP_C_LOW",   11),  # 11/1009 = 1.09018830...%
+    ("OVERREP_D_TINY",   5),  #  5/1009 = 0.49554014...%
+    ("OVERREP_E_EDGE",   2),  #  2/1009 = 0.19821605...%  (just above 0.1% threshold)
+]
+
+def random_seq(rng):
+    return "".join(rng.choice("ACGT") for _ in range(LEN))
+
+# Build the five overrepresented sequences. Use deterministic RNG, but reject
+# any candidate that collides with a previous sequence to keep counts exact.
+overrep_seqs = []
+seen = set()
+seq_rng = random.Random(20260426)
+for label, _count in OVERREP:
+    while True:
+        s = random_seq(seq_rng)
+        if s not in seen:
+            overrep_seqs.append(s)
+            seen.add(s)
+            break
+
+# Build background reads. Each must be unique AND not collide with any
+# overrepresented sequence (otherwise the percentages drift).
+n_overrep = sum(c for _, c in OVERREP)
+n_background = TOTAL - n_overrep
+assert n_background > 0
+background_seqs = []
+while len(background_seqs) < n_background:
+    s = random_seq(seq_rng)
+    if s in seen:
+        continue
+    seen.add(s)
+    background_seqs.append(s)
+
+# Assemble reads in a deterministic interleaved order so the output is stable.
+reads = []
+for s, (label, count) in zip(overrep_seqs, OVERREP):
+    for i in range(count):
+        reads.append((f"{label}_{i+1}", s))
+for i, s in enumerate(background_seqs):
+    reads.append((f"BACKGROUND_{i+1}", s))
+
+# Shuffle deterministically so overrepresented reads aren't all clustered.
+order_rng = random.Random(99)
+order_rng.shuffle(reads)
+assert len(reads) == TOTAL
+
+out_path = sys.argv[1]
+# mtime=0 makes the gzip header deterministic so re-running the generator
+# produces byte-identical output on any machine.
+with gzip.GzipFile(filename=out_path, mode="wb", mtime=0) as f:
+    for header, seq in reads:
+        f.write(f"@{header}\n{seq}\n+\n{QUAL}\n".encode("ascii"))
+
+print(f"wrote {out_path}: {TOTAL} reads, {LEN}bp, {len(OVERREP)} overrepresented sequences", file=sys.stderr)
+for label, count in OVERREP:
+    pct = count * 100 / TOTAL
+    print(f"  {label}: {count}/{TOTAL} = {pct}%", file=sys.stderr)

tests/equivalence/patches/realistic_default_adapter_content_svg.patch

@@ -0,0 +1,11 @@
+--- java/Images/adapter_content.svg
++++ rust/Images/adapter_content.svg
+@@ -17,7 +17,7 @@
+ <text x="380" y="30" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">% Adapter</text>
+ <line x1="30" y1="560" x2="790" y2="560" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+ <line x1="30" y1="560" x2="30" y2="40" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+-<text x="340" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Position in read (bp)</text>
++<text x="350" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Position in read (bp)</text>
+ <text x="30" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">1</text>
+ <rect width="20" height="520" x="50" y="40" style="fill:rgb(230,230,230);stroke:none"/>
+ <text x="50" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">2</text>

tests/equivalence/patches/realistic_default_duplication_levels_svg.patch

@@ -0,0 +1,22 @@
+--- java/Images/duplication_levels.svg
++++ rust/Images/duplication_levels.svg
+@@ -17,7 +17,7 @@
+ <text x="260" y="30" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Percent of seqs remaining if deduplicated 87.81%</text>
+ <line x1="30" y1="560" x2="790" y2="560" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+ <line x1="30" y1="560" x2="30" y2="40" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+-<text x="320" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Sequence Duplication Level</text>
++<text x="330" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Sequence Duplication Level</text>
+ <text x="50" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">1</text>
+ <rect width="50" height="520" x="80" y="40" style="fill:rgb(230,230,230);stroke:none"/>
+ <text x="100" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">2</text>
+@@ -68,7 +68,7 @@
+ <line x1="620" y1="560" x2="660" y2="560" stroke="rgb(136,34,85)" stroke-width="[SW]"/>
+ <line x1="660" y1="560" x2="710" y2="560" stroke="rgb(136,34,85)" stroke-width="[SW]"/>
+ <line x1="710" y1="560" x2="760" y2="560" stroke="rgb(136,34,85)" stroke-width="[SW]"/>
+-<rect width="120" height="23" x="670" y="40" style="fill:rgb(255,255,255);stroke:none"/>
+-<rect width="120" height="23" x="670" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
+-<text x="670" y="55" fill="rgb(136,34,85)" font-family="[FONT]" font-size="12">% Total sequences</text>
++<rect width="110" height="23" x="680" y="40" style="fill:rgb(255,255,255);stroke:none"/>
++<rect width="110" height="23" x="680" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
++<text x="680" y="55" fill="rgb(136,34,85)" font-family="[FONT]" font-size="12">% Total sequences</text>
+ </svg>

tests/equivalence/patches/realistic_default_per_base_n_content_svg.patch

@@ -0,0 +1,21 @@
+--- java/Images/per_base_n_content.svg
++++ rust/Images/per_base_n_content.svg
+@@ -17,7 +17,7 @@
+ <text x="330" y="30" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">N content across all bases</text>
+ <line x1="30" y1="560" x2="790" y2="560" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+ <line x1="30" y1="560" x2="30" y2="40" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+-<text x="340" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Position in read (bp)</text>
++<text x="350" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Position in read (bp)</text>
+ <text x="30" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">1</text>
+ <rect width="20" height="520" x="40" y="40" style="fill:rgb(230,230,230);stroke:none"/>
+ <text x="50" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">2</text>
+@@ -132,7 +132,7 @@
+ <line x1="730" y1="560" x2="740" y2="560" stroke="rgb(136,34,85)" stroke-width="[SW]"/>
+ <line x1="740" y1="560" x2="760" y2="560" stroke="rgb(136,34,85)" stroke-width="[SW]"/>
+ <line x1="760" y1="560" x2="770" y2="560" stroke="rgb(136,34,85)" stroke-width="[SW]"/>
+-<rect width="20" height="23" x="770" y="40" style="fill:rgb(255,255,255);stroke:none"/>
+-<rect width="20" height="23" x="770" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
++<rect width="20" height="23" x="760" y="40" style="fill:rgb(255,255,255);stroke:none"/>
++<rect width="20" height="23" x="760" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
+ <text x="770" y="55" fill="rgb(136,34,85)" font-family="[FONT]" font-size="12">%N</text>
+ </svg>

tests/equivalence/patches/realistic_default_per_base_quality_svg.patch

@@ -0,0 +1,20 @@
+--- java/Images/per_base_quality.svg
++++ rust/Images/per_base_quality.svg
+@@ -21,7 +21,7 @@
+ <text x="0" y="121" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">30</text>
+ <text x="0" y="91" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">32</text>
+ <text x="0" y="61" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">34</text>
+-<text x="250" y="30" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Quality scores across all bases (Illumina 1.5 encoding)</text>
++<text x="240" y="30" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Quality scores across all bases (Illumina 1.5 encoding)</text>
+ <rect width="20" height="297" x="20" y="263" style="fill:rgb(230,175,175);stroke:none"/>
+ <rect width="20" height="119" x="20" y="144" style="fill:rgb(230,215,175);stroke:none"/>
+ <rect width="20" height="104" x="20" y="40" style="fill:rgb(175,230,175);stroke:none"/>
+@@ -203,7 +203,7 @@
+ <rect width="20" height="104" x="760" y="40" style="fill:rgb(195,230,195);stroke:none"/>
+ <line x1="20" y1="560" x2="790" y2="560" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+ <line x1="20" y1="560" x2="20" y2="40" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+-<text x="340" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Position in read (bp)</text>
++<text x="350" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Position in read (bp)</text>
+ <rect width="10" height="0" x="20" y="427" style="fill:rgb(240,240,0);stroke:none"/>
+ <rect width="10" height="0" x="20" y="427" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(0,0,0)"/>
+ <line x1="30" y1="427" x2="30" y2="427" stroke="rgb(0,0,0)" stroke-width="[SW]"/>

tests/equivalence/patches/realistic_default_per_base_sequence_content_svg.patch

@@ -0,0 +1,22 @@
+--- java/Images/per_base_sequence_content.svg
++++ rust/Images/per_base_sequence_content.svg
+@@ -17,7 +17,7 @@
+ <text x="310" y="30" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Sequence content across all bases</text>
+ <line x1="30" y1="560" x2="790" y2="560" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+ <line x1="30" y1="560" x2="30" y2="40" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+-<text x="340" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Position in read (bp)</text>
++<text x="350" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Position in read (bp)</text>
+ <text x="30" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">1</text>
+ <rect width="20" height="520" x="40" y="40" style="fill:rgb(230,230,230);stroke:none"/>
+ <text x="50" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">2</text>
+@@ -279,8 +279,8 @@
+ <line x1="730" y1="443" x2="740" y2="446" stroke="rgb(221,204,119)" stroke-width="[SW]"/>
+ <line x1="740" y1="446" x2="760" y2="443" stroke="rgb(221,204,119)" stroke-width="[SW]"/>
+ <line x1="760" y1="443" x2="770" y2="448" stroke="rgb(221,204,119)" stroke-width="[SW]"/>
+-<rect width="20" height="83" x="770" y="40" style="fill:rgb(255,255,255);stroke:none"/>
+-<rect width="20" height="83" x="770" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
++<rect width="20" height="83" x="760" y="40" style="fill:rgb(255,255,255);stroke:none"/>
++<rect width="20" height="83" x="760" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
+ <text x="770" y="55" fill="rgb(136,34,85)" font-family="[FONT]" font-size="12">%T</text>
+ <text x="770" y="75" fill="rgb(51,34,136)" font-family="[FONT]" font-size="12">%C</text>
+ <text x="770" y="95" fill="rgb(17,119,51)" font-family="[FONT]" font-size="12">%A</text>

tests/equivalence/patches/realistic_default_per_sequence_gc_content_svg.patch

@@ -0,0 +1,22 @@
+--- java/Images/per_sequence_gc_content.svg
++++ rust/Images/per_sequence_gc_content.svg
+@@ -15,7 +15,7 @@
+ <text x="310" y="30" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">GC distribution over all sequences</text>
+ <line x1="30" y1="560" x2="790" y2="560" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+ <line x1="30" y1="560" x2="30" y2="40" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+-<text x="340" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Mean GC content (%)</text>
++<text x="350" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Mean GC content (%)</text>
+ <text x="30" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">0</text>
+ <rect width="10" height="520" x="40" y="40" style="fill:rgb(230,230,230);stroke:none"/>
+ <text x="40" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">2</text>
+@@ -303,8 +303,8 @@
+ <line x1="710" y1="560" x2="720" y2="560" stroke="rgb(51,34,136)" stroke-width="[SW]"/>
+ <line x1="720" y1="560" x2="720" y2="560" stroke="rgb(51,34,136)" stroke-width="[SW]"/>
+ <line x1="720" y1="560" x2="730" y2="560" stroke="rgb(51,34,136)" stroke-width="[SW]"/>
+-<rect width="160" height="43" x="640" y="40" style="fill:rgb(255,255,255);stroke:none"/>
+-<rect width="160" height="43" x="640" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
++<rect width="150" height="43" x="640" y="40" style="fill:rgb(255,255,255);stroke:none"/>
++<rect width="150" height="43" x="640" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
+ <text x="640" y="55" fill="rgb(136,34,85)" font-family="[FONT]" font-size="12">GC count per read</text>
+ <text x="640" y="75" fill="rgb(51,34,136)" font-family="[FONT]" font-size="12">Theoretical Distribution</text>
+ </svg>

tests/equivalence/patches/realistic_default_per_sequence_quality_svg.patch

@@ -0,0 +1,26 @@
+--- java/Images/per_sequence_quality.svg
++++ rust/Images/per_sequence_quality.svg
+@@ -9,10 +9,10 @@
+ <text x="0" y="257" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">600</text>
+ <text x="0" y="154" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">800</text>
+ <text x="0" y="51" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">1000</text>
+-<text x="290" y="30" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Quality score distribution over all sequences</text>
++<text x="280" y="30" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Quality score distribution over all sequences</text>
+ <line x1="40" y1="560" x2="790" y2="560" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+ <line x1="40" y1="560" x2="40" y2="40" stroke="rgb(0,0,0)" stroke-width="[SW]"/>
+-<text x="290" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Mean Sequence Quality (Phred Score)</text>
++<text x="300" y="595" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">Mean Sequence Quality (Phred Score)</text>
+ <text x="410" y="575" fill="rgb(0,0,0)" font-family="[FONT]" font-size="12">9</text>
+ <line x1="40" y1="560" x2="790" y2="560" stroke="rgb(180,180,180)" stroke-width="[SW]"/>
+ <line x1="40" y1="457" x2="790" y2="457" stroke="rgb(180,180,180)" stroke-width="[SW]"/>
+@@ -20,7 +20,7 @@
+ <line x1="40" y1="251" x2="790" y2="251" stroke="rgb(180,180,180)" stroke-width="[SW]"/>
+ <line x1="40" y1="148" x2="790" y2="148" stroke="rgb(180,180,180)" stroke-width="[SW]"/>
+ <line x1="40" y1="45" x2="790" y2="45" stroke="rgb(180,180,180)" stroke-width="[SW]"/>
+-<rect width="160" height="23" x="630" y="40" style="fill:rgb(255,255,255);stroke:none"/>
+-<rect width="160" height="23" x="630" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
+-<text x="630" y="55" fill="rgb(136,34,85)" font-family="[FONT]" font-size="12">Average Quality per read</text>
++<rect width="150" height="23" x="640" y="40" style="fill:rgb(255,255,255);stroke:none"/>
++<rect width="150" height="23" x="640" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
++<text x="640" y="55" fill="rgb(136,34,85)" font-family="[FONT]" font-size="12">Average Quality per read</text>
+ </svg>

tests/equivalence/patches/realistic_default_sequence_length_distribution_svg.patch

@@ -0,0 +1,13 @@
+--- java/Images/sequence_length_distribution.svg
++++ rust/Images/sequence_length_distribution.svg
+@@ -25,7 +25,7 @@
+ <line x1="40" y1="45" x2="790" y2="45" stroke="rgb(180,180,180)" stroke-width="[SW]"/>
+ <line x1="160" y1="560" x2="410" y2="40" stroke="rgb(136,34,85)" stroke-width="[SW]"/>
+ <line x1="410" y1="40" x2="660" y2="560" stroke="rgb(136,34,85)" stroke-width="[SW]"/>
+-<rect width="110" height="23" x="680" y="40" style="fill:rgb(255,255,255);stroke:none"/>
+-<rect width="110" height="23" x="680" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
+-<text x="680" y="55" fill="rgb(136,34,85)" font-family="[FONT]" font-size="12">Sequence Length</text>
++<rect width="100" height="23" x="680" y="40" style="fill:rgb(255,255,255);stroke:none"/>
++<rect width="100" height="23" x="680" y="40" rx="0" ry="0" style="fill:none;stroke-width:1;stroke:rgb(192,192,192)"/>
++<text x="690" y="55" fill="rgb(136,34,85)" font-family="[FONT]" font-size="12">Sequence Length</text>
+ </svg>