Bump version to 3.15.0 and write release notes

README.md

@@ -1,6 +1,6 @@
 [![CI](https://github.com/amrali-eg/EncodingChecker/actions/workflows/ci.yml/badge.svg)](https://github.com/amrali-eg/EncodingChecker/actions/workflows/ci.yml)
 
-# EncodingChecker v3.14.5
+# EncodingChecker v3.15.0
 
 EncodingChecker is a Windows tool for finding, checking, and safely converting text-file encodings. Use the GUI for everyday work or the command line for repeatable jobs.
 

docs/RELEASE-NOTES-v3.15.0.md

@@ -0,0 +1,91 @@
+# EncodingChecker v3.15.0
+
+Plain ASCII text is now recognised as already being UTF-8, so repeating a conversion no
+longer replaces the backup of a file's original. Seven smaller fixes tighten what EC
+refuses, what it saves, and what the window tells you. This release changes conversion
+policy and advances the conversion semantics version, so plans saved by any 3.14.x
+release must be made again.
+
+## ASCII is already UTF-8
+
+ASCII bytes are identical in UTF-8 without a byte-order mark, but EC planned them as a
+conversion. A UTF-16 file of English text converts to ASCII bytes, so running
+`-Target utf-8 -Backup` a second time rewrote the file to the same bytes and replaced its
+`.bak`, which held the UTF-16 original, with a copy of the converted file. The recovery
+record describing the original went with it. The text was never at risk; the only copy of
+the original bytes was. `-FailOnChanges` also reported a change on every run.
+
+An ASCII file with a UTF-8 target, both without a BOM, is now **Unchanged**, with the
+explanation "ASCII is already valid UTF-8 without a BOM." The whole file is still
+validated, so a non-ASCII byte past the 64 KiB detection sample is reported as an error,
+now before any backup is made. A `utf-8-bom` target still converts ASCII.
+
+`-Validate` follows the same rule: `-Validate "utf-8"` accepts an ASCII file, and a list
+allowing only `utf-8-bom` still rejects it.
+
+## Other fixes
+
+- **A source choice that contradicts reliable detection is refused even when it is also
+  the target.** `-From windows-1252 -Target windows-1252` on a UTF-8 file used to report
+  it as already windows-1252 with exit 0. It is now refused, exit 5, as any other
+  contradicting choice is.
+- **A source-choice warning stays with a file left unchanged.** Choosing a source for
+  BOM-less UTF-16 or UTF-32 that EC cannot confirm is flagged whether or not the file then
+  needs rewriting, in reports and in the review window.
+- **A file that changes after its conversion was decided keeps its earlier backup.** The
+  backup copy now checks the bytes it copies against the approved snapshot. Before, a file
+  changed after a plan's stale check but before its own turn was correctly refused, yet
+  its `.bak` had already been replaced with the changed bytes and its recovery record
+  deleted.
+- **Read-only and linked output files are refused, not replaced.** This now covers plans,
+  journals, CLI reports, settings and the GUI's journal export, not only the GUI's text
+  and CSV exports. The CLI checks `-Plan`, `-Journal` and `-Report` before any file
+  changes and exits 3. A read-only or linked `Settings.xml` is no longer replaced, so
+  preferences are not saved while it stays that way.
+- **Cancelling a review undoes the source choices made in it.** Before, the next review
+  showed those files as ready with a source nobody chose in it. Choices made before the
+  review opened are kept.
+- **The window's Validate status says how many files it checked.** It read "0 files do
+  not have the correct encoding" both when every file passed and when no file matched.
+  It now reads "No matching files were examined", "Checked N files: all valid", or the
+  count that failed, with any files that could not be read.
+- **Result rows show this run's state and reason.** A row left unchanged or skipped no
+  longer keeps an earlier run's icon, and hovering a row shows why it ended as it did.
+
+## Compatibility
+
+**Conversion semantics move from 7 to 8. Plans written by v3.14.x or earlier are
+refused.** Re-run `-Plan` and review the result. The plan schema stays at 6 and the journal
+schema at 6, so other readers of these files are unaffected.
+
+Scripts may notice:
+
+- `-Target utf-8` reports ASCII files as `Unchanged` instead of `Converted`, and
+  `-FailOnChanges` passes on them.
+- `-Validate "utf-8"` passes ASCII files.
+- Some explicit source choices that used to end as `Unchanged` (exit 0) or as a
+  validation error (exit 3) are now refused (exit 5).
+- A read-only or linked `-Plan`, `-Journal` or `-Report` file now stops the run with
+  exit 3 before any file changes.
+- The window's Validate status line has new wording.
+
+The CSV and JSON formats are unchanged.
+
+## Verification
+
+- 980 tests pass, none skipped; release build with no warnings.
+- `docs/Test-DefectBacklog.ps1` passes: 87 findings, 75 fixed, 8 open.
+- The **Shared Unicode detector parity** check is unaffected: neither shared detector
+  file changed.
+- This release changes conversion policy, so the four-corpus audit is required. It runs
+  against this exact commit, and its results are recorded in `SAFETY-AUDIT.md` after
+  tagging.
+- **The executables are not code-signed.** The signing secrets are not configured, so
+  the release workflow skips signing. Shipping unsigned is a decision for this release,
+  not a passed check.
+
+## Known open issues
+
+Recorded in `DEFECT-BACKLOG.md`: a preview's CSV report still calls files `Converted`
+(BL-42), and a file name with an unpaired UTF-16 surrogate cannot be planned or backed up
+(BL-43). Neither changes a file's bytes.

sources/EncodingChecker/Properties/AssemblyInfo.cs

@@ -41,5 +41,5 @@
 // You can specify all the values, or you can default the Build and Revision Numbers
 // by using the '*' as shown below:
 // [assembly: AssemblyVersion("1.0.*")]
-[assembly: AssemblyVersion("3.14.5.0")]
-[assembly: AssemblyFileVersion("3.14.5.0")]
+[assembly: AssemblyVersion("3.15.0.0")]
+[assembly: AssemblyFileVersion("3.15.0.0")]