Fix four TLE reader issues in satTle2elem

docs/source/Support/bskKnownIssuesSnippets/1676-tle-blank-designator.rst

@@ -0,0 +1,9 @@
+- In versions 2.8.28 through 2.12.2, :ref:`tleHandling` ``satTle2elem()`` raised
+  ``ValueError`` for a whole file when any set had a blank international designator
+  (columns 10-17 of line 1), as CelesTrak's analyst-object TLEs do. Such sets now load with
+  ``launchDate`` left as ``None``; ``launchNo`` and ``pol`` retain their defaults of
+  ``1`` and ``"A"``, which do not represent known launch metadata for these objects.
+  Upgrade to a version containing this fix or backport the reader change to read
+  these sets. Until then, omit the complete affected sets to read the rest of the file.
+  The :ref:`ommHandling` reader was introduced during the 2.13 beta cycle and is
+  unavailable in the affected releases.

docs/source/Support/bskKnownIssuesSnippets/1676-tle-ephemeris-type-4.rst

@@ -0,0 +1,11 @@
+- In versions 2.8.28 through 2.12.2, :ref:`tleHandling` ``satTle2elem()`` converted
+  TLEs with ephemeris type 4 (SGP4-XP) using plain SGP4,
+  which can produce incorrect states because those sets carry BTERM and AGOM in
+  the B* and second-derivative columns. Unsupported sets now raise ``ValueError``
+  by default. With ``on_error="skip"``, they are skipped with a printed warning,
+  while other readable sets are returned. If every attempted set fails in skip
+  mode, the first error is raised; a file containing only otherwise valid type-4
+  sets raises ``ValueError`` in either mode. With an affected version, check
+  column 63 of line 1 and exclude complete sets unless that field is blank,
+  ``0`` or ``2``.
+  Changing a type-4 field to ``0`` or ``2`` does not convert SGP4-XP data to SGP4.

docs/source/Support/bskKnownIssuesSnippets/1676-tle-name-lines.rst

@@ -0,0 +1,15 @@
+- In versions 2.8.28 through 2.12.2, :ref:`tleHandling` ``satTle2elem()`` raised
+  ``ValueError`` for a whole 3LE file when any name line started with ``1`` or ``2``
+  (for example ``1KUNS-PF``), because such lines were counted as element lines. Issue
+  #1676 is fixed by recognizing complete record layouts before counting element-line
+  prefixes, preserving names such as ``1 TEST`` and ``2 TEST`` and their associated
+  satellites, including files with trailing blank lines. These prefixes are also
+  accepted in optional 2LE file titles. Recognizable truncated element headers are
+  rejected as possible orphaned lines rather than consumed as names or file titles.
+  This includes longer fragments with malformed catalog fields when later field
+  structure still identifies them as element lines, and shortened headers when
+  intact sequences of element fields identify columns shifted by missing characters.
+  Prefix an intentional element-like title with ``0`` followed by a space to
+  disambiguate it; for a satellite name, that prefix is retained in ``satName``.
+  With an earlier version, rename satellites or file titles starting with ``1`` or
+  ``2`` before reading the file.

docs/source/Support/bskKnownIssuesSnippets/1676-tle-skip-bad-set.rst

@@ -0,0 +1,9 @@
+- In versions 2.8.28 through 2.12.2, :ref:`tleHandling` ``satTle2elem()`` failed for
+  the whole file when any one set could not be read (for example a bad checksum or
+  a malformed field), so no valid sets were returned either. This remains the
+  default behavior for compatibility. Call ``satTle2elem(path, on_error="skip")``
+  to recover readable sets: complete sets that raise validation or arithmetic
+  errors, including epoch overflow and division by zero, are skipped with a
+  printed warning. If every attempted set fails in skip mode, the first set's
+  original error is raised. File-level format checks, including orphaned element
+  lines, can still reject the whole file in either mode.

docs/source/Support/bskReleaseNotesSnippets/1676-tle-blank-designator.rst

@@ -0,0 +1 @@
+- :ref:`tleHandling` ``satTle2elem()`` now reads element sets with a blank international designator, such as CelesTrak's analyst objects, leaving ``launchDate`` as ``None``.

docs/source/Support/bskReleaseNotesSnippets/1676-tle-ephemeris-type.rst

@@ -0,0 +1 @@
+- :ref:`tleHandling` ``satTle2elem()`` accepts a blank ephemeris-type field, ``0`` or ``2`` for SGP4 conversion. Other types, including ``4`` (SGP4-XP), raise ``ValueError`` by default. With ``on_error="skip"``, unsupported sets are skipped with a printed warning; if every attempted set fails, the first error is raised.

docs/source/Support/bskReleaseNotesSnippets/1676-tle-name-lines.rst

@@ -0,0 +1,2 @@
+- :ref:`tleHandling` ``satTle2elem()`` now preserves satellite names and records in 3LE files whose names start with ``1`` or ``2``, including ``1 TEST`` and ``2 TEST``, accepts these prefixes in optional 2LE file titles, and ignores trailing blank lines.
+- :ref:`tleHandling` rejects recognizable orphaned element lines, including truncated lines, shortened headers with missing characters, and longer fragments with malformed catalog fields, instead of consuming them as names or file titles. Prefix an intentional element-like title with ``0`` followed by a space to disambiguate it.

docs/source/Support/bskReleaseNotesSnippets/1676-tle-skip-bad-set.rst

@@ -0,0 +1 @@
+- :ref:`tleHandling` ``satTle2elem()`` now accepts the keyword-only option ``on_error="skip"`` to skip complete sets that fail validation or numerical conversion, including epoch overflow and division by zero, with a printed warning and return the readable sets. The default, ``on_error="raise"``, preserves the existing behavior of raising the first record error immediately. In skip mode, the first error is raised if every attempted set fails. File-level format checks can still reject the entire file in either mode.