1
2 09/21/87 mrpg
3 Known errors in the current release of mrpg.
4 # Associated TR's
5 Description
6
7 69 phx16531
8 H_F_common.I_page is improperly declared, causing PL/I compilation
9 errors due to data mismatch.
10
11 68 phx16531
12 MRPG fails with various faults when compiling
13 a program with >10 nested IF statements.
14
15 58 phx12612
16 There is a limitation of 100 on the number of variables which may be
17 DCL'ed in an MRPG program. There is an expense involved in having this
18 limit be "very" large and 100 was picked as a compromise between the
19 expense and the usefullness. number of DCL's in a program.
20
21 51 phx09474
22 MRPG seems to have a bug in the interaction of control breaks with page
23 headings.
24
25 42
26 When a PAGEHEAD is defined with an absolute line number less than 4
27 when output is not going to user_output and then is followed up by
28 relative line references, extra blank lines are generated until it
29 thinks it has made it past line 4.
30
31 41
32 When a level 4 value includes the CHARn attribute, if the data given
33 is longer than n then improper formating occurs.
34
35 37
36 The action of TRANSFORM when the key supplied is not in table is not
37 properly documented.
38
39 36 phx01889
40 The SWITCH keyword is not documented clearly.
41
42 32
43 Varying input/parameter fields passed to set/table procedures gives a
44 warning due to data-type mismatch.
45
46 30 phx03244
47 DETAILHEAD MAXLINE does not work.
48
49 28
50 Some combinations of empty detail lines and headers/footers causes
51 improper page layout. P_line needs rel_lineabs_line parameter
52
53 26
54 MRPG reports need to make a better check on record length vs data field
55 definitions.
56
57 24
58 The parse_failed message is misleading when the problem is a missing
59 DETAIL keyword.
60
61 23
62 The PAUSE option on a line generates and compiles without complaint,
63 but it does not pause.
64
65 19
66 When a SET is declared and then not used, MRPG generates the
67 corresponding internal procedure. Since this procedure is not
68 referenced, PL/I gives a warning.
69
70 10
71 A warning is issued if a comment contains an imbedded /*, but the line
72 number may be wrong.
73
74 7
75 When the user is HOLDing data and the amount gets very large,
76 record_quota_overflow can occur in the process directory.
77
78 5
79 When the user leaves out a necessary SP delimiter the parse sometimes
80 fails without being very informative.