maintenance: anonymise the named science target in the WFC3-IR hole regression

autoreduce/instruments/wfc3_ir.py

@@ -34,11 +34,11 @@
         # blob bit plus hot pixels. Blobs are detector-fixed IR channel
         # features that calwf3 flags but does not remove; rejecting them on
         # snapshot data with tiny dithers punches structured zero-coverage
-        # holes in the mosaic (PJ011646, 5 exposures, a 123-px hole at
-        # r = 5.3"). The IR rows are the reason this is a table and not a
-        # pair: the two bits columns DIFFER at N = 2-3, where the separate
-        # (median-building) drizzle still keeps every bit while the final
-        # drizzle already drops to 528.
+        # holes in the mosaic (program 14653, F160W, 5 exposures, a
+        # 123-px hole at r = 5.3"). The IR rows are the reason this is a
+        # table and not a pair: the two bits columns DIFFER at N = 2-3,
+        # where the separate (median-building) drizzle still keeps every
+        # bit while the final drizzle already drops to 528.
         dq_bits_rows=((1, 65535, 65535), (2, 65535, 528), (4, 528, 528)),
     )
 )

docs/design/wfc3.md

@@ -49,10 +49,11 @@ which `calwf3` *flags* but does not remove. They are fixed in detector
 coordinates, so a dither pattern of a few pixels moves them barely at all: the
 same detector pixels are flagged in every exposure of the visit. Reject them
 and the mosaic gets a structured **zero-coverage hole** rather than the
-speckle that a large dither would produce. PJ011646 (program 14653, F160W,
-5 exposures, 2–6 px dithers) failed packaging on exactly this — a single
-123-px hole at r = 5.3″, DQ 512 at the same detector pixels in all five
-exposures — while a trusted external reduction of the same data has none.
+speckle that a large dither would produce. An HST program 14653 dataset
+(F160W, 5 exposures, 2–6 px dithers) failed packaging on exactly this — a
+single 123-px hole at r = 5.3″, DQ 512 at the same detector pixels in all
+five exposures — while a trusted external reduction of the same data has
+none.
 STScI's own MDRIZTAB passes the blob bit, so under standard practice that hole
 could not have occurred.
 

test_autoreduce/test_target_and_instruments.py

@@ -154,10 +154,11 @@ def test_wfc3_ir_rows_differ_between_the_two_columns(self):
         assert self._bits("wfc3_ir", 4) == (528, 528)
         assert 528 == 512 | 16
 
-    def test_pj011646_would_not_have_holed(self):
-        # The regression this leg exists for: PJ011646 was F160W with five
-        # exposures, so it lands on the numimages >= 4 row, where STScI
-        # passes exactly the blob bit (512) that punched the 123-px hole.
+    def test_five_exposure_f160w_mosaic_would_not_have_holed(self):
+        # The regression this leg exists for: an HST program 14653 F160W
+        # mosaic with five exposures lands on the numimages >= 4 row, where
+        # STScI passes exactly the blob bit (512) that punched the 123-px
+        # hole.
         _, final_bits = self._bits("wfc3_ir", 5)
         assert final_bits & 512