Questions about chacking diffs

[KZiemian]

@rmhsilva make great job making diffs between CLHS and already done CL UltraSpec. Now we must check this diffs I find all mistekes in CLUS. This is some work to do, I probably can do that, but I need some help with things that I don't understand. I just need to know, if I need to marke things like that or not.

Current state of work is here https://github.com/KZiemian/clus-data/tree/master/diffs

I apologize for my english.

##############################

1. File cl:functions:acosh.diff

A) Maybe wrong TeX format of mathematical functions. Example lines 39-49.

-Hyperbolic sine         (e^x-e^-x)/2
-Hyperbolic cosine       (e^x+e^-x)/2
-Hyperbolic tangent      (e^x-e^-x)/(e^x+e^-x)
-Hyperbolic arc sine     log  (x+sqrt(1+x^2))
-Hyperbolic arc cosine   2 log  (sqrt((x+1)/2) + sqrt((x-1)/2))
+ Hyperbolic sine  (ex - e-x) / 2
+ Hyperbolic cosine  (ex + e-x) / 2
+ Hyperbolic tangent  (ex - e-x) / (ex + e-x)
+ Hyperbolic arc sine  log(x + sqrt(1 + x2))
+ Hyperbolic arc cosine  2 * log(sqrt((x + 1) / 2) + sqrt((x - 1) / 2))
 Hyperbolic arc tangent  (log  (1+x) - log (1-x))/2

B) New style of section references. Example lines 100-103.

-log, sqrt, Section 12.1.3.3 (Rule of Float Substitutability)
+ Function LOG
+ Function SQRT
+ {\secref\FloatSubstitutability}

############################## 2) File cl:declarations:notinline.diff. Do upacase names in orginal sholud stay that way or not?

Example lines 48-51.

- ;; The globally defined function DISPATCH should be open-coded,
- ;; if the implementation supports inlining, unless a NOTINLINE
- ;; declaration overrides this effect.
+The globally defined function dispatch should be open-coded, if the implementation supports inlining, unless a notinline declaration overrides this effect.

############################## 3) Often <TT> line is missing. Notice this or not?

Example from file cl:constant_variables:boole-1.diff, line 9-13.

- (boole boole-eqv 17 15) =>  -31

-<TT>
+(boole boole-ior 1 16)

############################## 4) Similary to that, sometimes <A> is missing.

Example from cl:declarations:dynamic-extent.diff, lines 7-8.

-(dynamic-extent [[var* | (function<A> fn)*]])
+ (dynamic-extent [[var* | (function fn)*]])

############################## 5) Changed code style.

Example. File cl:functions:adjoin.diff, lines 31-47

- (setq slist '()) =>  NIL
- (adjoin 'a slist) =>  (A)
- slist =>  NIL
- (setq slist (adjoin '(test-item 1) slist)) =>  ((TEST-ITEM 1))
- (adjoin '(test-item 1) slist) =>  ((TEST-ITEM 1) (TEST-ITEM 1))
- (adjoin '(test-item 1) slist test 'equal) =>  ((TEST-ITEM 1))
- (adjoin '(new-test-item 1) slist key #'cadr) =>  ((TEST-ITEM 1))
- (adjoin '(new-test-item 1) slist) =>  ((NEW-TEST-ITEM 1) (TEST-ITEM 1))

-<TT>
-Affected By None.
+(defparameter *slist* '())
+→
+*SLIST*
+(adjoin 'a *slist*)
+→

############################## 6) Changed style of symbols notations.

Examples from file cl:functions:adjust-array.diff

A) Quotations added Example lines 42-43.

-A is not displaced before, but is displaced to C after the call
+''A'' is not displaced before, but is displaced to ''C'' after the call

B) Changed citation style. Example line 63.

-<TT> A gets a new ``data region,''
Linia 65.
+A gets a new "data region,"

############################## ############################## ############################## Some file were missed by diff and they diffs are empty. This is a list of that files NOT checked by us to this moment.

cl:functions:hash-table.p.diff cl:types:restart.diff cl:types:satisfies.diff cl:types:standard-class.p.diff cl:types:standard-object.diff cl:types:storage-condition.diff cl:functions:setf-table.p.diff cl:functions:setf-class-name.diff cl:types:control-error.diff cl:types:division-by-zero.diff cl:types:floating-point-inexact.diff cl:types:floating-point-invalid-operation.diff cl:types:floating-point-overflow.diff cl:types:floating-point-underflow.diff cl:types:generic-function.diff cl:types:method-combination.diff cl:types:program-error.diff cl:types:restart.diff cl:types:satisfies.diff cl:types:standard-class.diff cl:types:standard-object.diff cl:types:sotrage-condition.diff

[rmhsilva]

@KZiemian this is great, thank you for your work! I'm busy this weekend but will get back on to CLUS next week.

The files that were missed we will probably have to check manually, as they did not neatly map between CLUS pages and the spec (which is fine).

[KZiemian]

@rmhsilva I and @phoe probably meet and discuss this problem at Wednesday, so don't bother about this to Thursday. After that I will post problems there are still open.

[rmhsilva]

@KZiemian ok, thanks.

[KZiemian]

@rmhsilva I talk with @phoe today and my problems are solved. I will keep this issues open, because other complications may came along the way.

Also @phoe sugest that this empyt filles are identical in both versions. Can you check if that is true?

[rmhsilva]

Good to hear some problems are solved. The empty diffs indicate that the originals are identical, correct.

[KZiemian]

@phoe Should we check "See Also" and another things that are below "Examples"?

[phoe]

@KZiemian What do you mean by things that are below "Examples"? See Also should contain only links to other parts of the standard - it should be checked.

[KZiemian]

@phoe I mean that in files typical there is description parts, after that there are "Examples" after that are "See Also", "Notes" and maybe something else. In past I think that everything after "Examples" are things outside standard, so I just not care about it. No I am rechecking all diffs done looking only on parts below "Examples". Fortunatlny after 10 or so files, not one need additional adnotations.

Thank for help, I go back to diffs.

[KZiemian]

@phoe One more question. What mean "\ No newline at end of file" at the end of the file? Like in cl:declarations:special.diff:

\ No newline at end of file +Symbol DECLARE, Macro DEFPARAMETER, Macro DEFVAR \ No newline at end of file

Noting that?

[phoe]

"\ No newline at end of file"

Ignore them.

[KZiemian]

@phoe Ok. I check in all files text after "Examples" everything looks right.

I find in last hour another phenomenon, line doubeling. Like in cl:functions:alpha-char-p.diff Arguments and Values

@@ -6,30 +10,75 @@

Arguments and Values

I don't think that is anything that that side effect of diff, but if note, we should recheck diffs looking for that.

[KZiemian]

Some information.

For all files checked +content is CLUS; -content is CLHS. @phoe does that help? @rmhsilva can we assume that in all files is that way?

In 99% of situations in CLUS exponent signs are missing: CLHS 2^x CLUS 2x.

[KZiemian]

Another iritaiting techincal problem. After downloading diffs on my hard driver in Ubuntu OS every colon was replaced by undersocre, so now when I make a commit diff are not updatet but dubled under another name. Anyone have a easy solution to that? Because now I have to much on my head to thinking about that.