Fix html/other sourcing; re-scramble Makefile

8 files changed, 283 insertions, 155 deletions

@@ -1,64 +1,51 @@
-NAME = jimmy
-
-LIBS = read emit util
-
-BUILD = $(PWD)/build
-SRC = $(PWD)/src
-TESTS = $(PWD)/tests
+# Automatically generated by scramble
 
+NAME = jimmy
 CSC = /usr/bin/csc
+CSC_OPTIONS = -setup-mode -host -I $(PWD) -C -I$(PWD)
+CSC_LIB_OPTIONS = -D compiling-extension -emit-all-import-libraries -dynamic -regenerate-import-libraries
+CSC_OPTIONS_EXTRA = -X utf8 -X module-declarations
 CSI = /usr/bin/csi
-CSC_OPTIONS = \
-        -setup-mode \
-        -host \
-        -D compiling-extension \
-        -emit-all-import-libraries \
-        -dynamic \
-        -regenerate-import-libraries \
-        -I $(SRC) \
-        -C -I$(SRC)
-
-CSC_OPTIONS_EXTRA = \
-        -X utf8 \
-        -X module-declarations
-
-## Library dependency graph
-# here's a convenience macro
-lib = $(BUILD)/$(NAME).$(1).so
-# and another
-src = $(SRC)/$(1).scm
-
-LIBS_ = $(foreach l,$(LIBS),$(call lib,$(l)))
-
-## Phonies
-
-.PHONY: build test clean
-build: $(LIBS_)
-        -mv *.import.scm build/
-
-test: build
-        cd $(BUILD) && \
-        $(CSI) -setup-mode -s $(TESTS)/run.scm $(NAME)
+BUILD = $(PWD)/build
+TESTS = $(PWD)/tests
+TEST_ENV = env BUILD=$(BUILD) TESTS=$(TESTS)
+TEST_ENV_EXTRA = TEST_USE_ANSI=0
 
+.PHONY: all test clean install uninstall
+all: build/jimmy.util.so build/jimmy.read.so build/jimmy.emit.so build/jimmy.html.so build/jimmy.wrap.so
+test: all
+        cd $(BUILD) && $(TEST_ENV) $(TEST_ENV_EXTRA) $(CSI) -setup-mode -s $(TESTS)/run.scm $(NAME)
 clean:
-        -rm -rf $(BUILD)
-
+        -rm -rf $(BUILD) *.build.sh *.install.sh $(NAME) *.import.scm *.so *.link *.static.o
 install:
         chicken-install -s
-
 uninstall:
         chicken-uninstall -s
 
-# Scm -> So
-
-$(BUILD)/$(NAME).%.so: src/%.scm
-        mkdir -p "$(dir $@)"
-        $(CSC) $(CSC_OPTIONS) $(CSC_OPTIONS_EXTRA) $< -o $@
-
-# Libraries!
-$(call lib,util): $(call src,util)
+# jimmy
 
-$(call lib,read): $(call src,read) $(call lib,util)
-$(call lib,emit): $(call src,emit) $(call lib,util)
-
-# Program!
+build/jimmy.util.so: src/util.scm
+        @mkdir -p $(BUILD)
+        cd $(BUILD) && \
+        $(CSC) $(CSC_OPTIONS) $(CSC_LIB_OPTIONS) $(CSC_OPTIONS_EXTRA) ../$< -o $(@F)
+        @test -f jimmy.util.import.scm &&mv jimmy.util.import.scm $(BUILD)/||true
+build/jimmy.read.so: src/read.scm src/util.scm
+        @mkdir -p $(BUILD)
+        cd $(BUILD) && \
+        $(CSC) $(CSC_OPTIONS) $(CSC_LIB_OPTIONS) $(CSC_OPTIONS_EXTRA) ../$< -o $(@F)
+        @test -f jimmy.read.import.scm &&mv jimmy.read.import.scm $(BUILD)/||true
+build/jimmy.emit.so: src/emit.scm src/util.scm
+        @mkdir -p $(BUILD)
+        cd $(BUILD) && \
+        $(CSC) $(CSC_OPTIONS) $(CSC_LIB_OPTIONS) $(CSC_OPTIONS_EXTRA) ../$< -o $(@F)
+        @test -f jimmy.emit.import.scm &&mv jimmy.emit.import.scm $(BUILD)/||true
+build/jimmy.html.so: src/html.scm src/util.scm src/emit.scm
+        @mkdir -p $(BUILD)
+        cd $(BUILD) && \
+        $(CSC) $(CSC_OPTIONS) $(CSC_LIB_OPTIONS) $(CSC_OPTIONS_EXTRA) ../$< -o $(@F)
+        @test -f jimmy.html.import.scm &&mv jimmy.html.import.scm $(BUILD)/||true
+build/jimmy.wrap.so: src/wrap.scm src/util.scm src/emit.scm
+        @mkdir -p $(BUILD)
+        cd $(BUILD) && \
+        $(CSC) $(CSC_OPTIONS) $(CSC_LIB_OPTIONS) $(CSC_OPTIONS_EXTRA) ../$< -o $(@F)
+        @test -f jimmy.wrap.import.scm &&mv jimmy.wrap.import.scm $(BUILD)/||true
@@ -7,11 +7,24 @@
  (component-options
   (csc-options -X utf8 -X module-declarations))
  (components
+  ;; Utility library
   (extension jimmy.util
              (source src/util.scm))
+  ;; Read gemini files into internal format
   (extension jimmy.read
              (source src/read.scm)
              (component-dependencies jimmy.util))
+  ;; Emit the output format (includes gemini)
   (extension jimmy.emit
              (source src/emit.scm)
-             (component-dependencies jimmy.util))))
+             (component-dependencies jimmy.util))
+  ;; Emit HTML -- import this *after* emit (is this the best way?)
+  (extension jimmy.html
+             (source src/html.scm)
+             (component-dependencies jimmy.util
+                                     jimmy.emit))
+  ;; Wrap output in templates
+  (extension jimmy.wrap
+             (source src/wrap.scm)
+             (component-dependencies jimmy.util
+                                     jimmy.emit))))
@@ -1,3 +1,5 @@
+(declare (module (jimmy emit)))
+
 (import scheme (chicken base)
         (chicken format)
         (chicken irregex)
@@ -9,41 +11,36 @@
   (for-each display (map format-stanza doc)))
 
 (define-public formats
-  ;;; (TYPE (line . LINE-FMT) (stanza . STANZA-FMT) (inline . INLINE-FMT))
-  '((para (line . "~A")
-          (stanza . "~A~%~%"))
-    (verb (line . "~A~%")
-          (stanza . "```~%~A```~%~%"))
-    (link (line . "=> ~A ~A~%")         ; Note: link has 2 format arguments
-          (stanza . "~A~%")
-          (inline . "~%=> ~A ~A~%"))
-    (list (line . "* ~A~%")
-          (stanza . "~A~%"))
-    (quot (line . "~A")
-          (stanza . "> ~A~%~%"))
-    (hdr1 (line . "# ~A~%")
-          (stanza . "~A~%"))
-    (hdr2 (line . "## ~A~%")
-          (stanza . "~A~%"))
-    (hdr3 (line . "### ~A~%")
-          (stanza . "~A~%"))
-    (meta (line . "")
-          (stanza . ""))
-    (default
-      (line . "~A")
-      (stanza . "~A~%~%"))))
+  (make-parameter
+   ;; (TYPE (line . LINE-FMT) (stanza . STANZA-FMT) (inline . INLINE-FMT))
+   '((para (line . "~A")
+           (stanza . "~A~%~%"))
+     (verb (line . "~A~%")
+           (stanza . "```~%~A```~%~%"))
+     (link (line . "=> ~A ~A~%")         ; Note: link has 2 format arguments
+           (stanza . "~A~%")
+           (inline . "~%=> ~A ~A~%"))
+     (list (line . "* ~A~%")
+           (stanza . "~A~%"))
+     (quot (line . "~A")
+           (stanza . "> ~A~%~%"))
+     (hdr1 (line . "# ~A~%")
+           (stanza . "~A~%"))
+     (hdr2 (line . "## ~A~%")
+           (stanza . "~A~%"))
+     (hdr3 (line . "### ~A~%")
+           (stanza . "~A~%")))))
 
 (define-public filters
-  ;;; (TYPE (line . LINE-FILTER) (stanza . STANZA-FILTER))
-  ;; line-filter : (lambda (list-of-strs) ...) -> list-of-strs (for format)
-  ;; stanza-filter : (lambda (list-of-strs) ...) -> str
-  `((verb (line . ,identity)
-          (stanza . ,(lambda (lines) (apply string-append lines))))
-    (default
-      (line . ,identity)
-      (stanza . ,(lambda (lines)
-                   (irregex-replace/all '(: bol (* space))
-                                        (string-join lines) ""))))))
+  (make-parameter
+   ;; (TYPE (line . LINE-FILTER) (stanza . STANZA-FILTER))
+   ;; line-filter : (lambda (list-of-strs) ...) -> list-of-strs (for format)
+   ;; stanza-filter : (lambda (list-of-strs) ...) -> str
+   `((verb (line . ,identity)
+           (stanza . ,join-lines))
+     (default
+       (line . ,identity)
+       (stanza . ,flush-lines-left)))))
 
 (define (format-line line el)
   (cond
@@ -72,8 +69,10 @@
       (and (eq? scope 'inline)
            (alist-walk alist 'default 'line))))
 
-(define (get-format el scope) (get-from formats el scope))
-(define (get-filter el scope) (get-from filters el scope))
+(define (get-format el scope)
+  (or (get-from (formats) el scope)
+      ""))
+(define (get-filter el scope) (get-from (filters) el scope))
 
 (define (sprintf* fmt lis)
   (let loop ((num (length (irregex-extract "~[aA]" fmt)))
@@ -1,3 +1,61 @@
 (declare (module (jimmy html)))
 
+(import scheme (chicken base)
+        (chicken irregex)
+        (jimmy emit)
+        (jimmy util))
 
+(define (escape-entities s)
+  (irregex-replace/all "[&<>]" s
+                       (lambda (m)
+                         (let ((c (irregex-match-substring m)))
+                           (cond
+                            ((equal? c "&") "&amp;")
+                            ((equal? c "<") "&lt;")
+                            ((equal? c ">") "&gt;"))))))
+
+(define (add-inline-markup s)
+  (define (char->tag ch tag)
+    (lambda (s)
+      (irregex-replace/all `(: ,ch ($ (* (~ ,ch))) ,ch) s
+                           "<" tag ">" 1 "</" tag ">")))
+
+  ((o (char->tag "*" "b")
+      (char->tag "_" "i")
+      (char->tag "`" "code")) s))
+
+(formats
+  '((para (line . "~a~%")
+          (stanza . "<p>~% ~a</p>~%"))
+    (verb (line . "~a~%")
+          (stanza . "<pre><code>~a</code></pre>~%"))
+    (link (line . "<li><a href=\"~a\">~a</a></li>~%")
+          (stanza . "<ul>~% ~a</ul>~%")
+          (inline . "<a href=\"~a\">~a</a>~%"))
+    (list (line . "<li>~a</li>~%")
+          (stanza . "<ul>~% ~a</ul>~%"))
+    (quot (line . "~a~%")
+          (stanza . "<blockquote>~% ~a</blockquote>~%"))
+    (hdr1 (line . "~a")
+          (stanza . "<h1>~a</h1>~%"))
+    (hdr2 (line . "~a")
+          (stanza . "<h2>~a</h2>~%"))
+    (hdr3 (line . "~a")
+          (stanza . "<h3>~a</h3>~%"))))
+
+(filters
+  `((verb (line . ,identity)
+          (stanza . ,join-lines))
+    (link (line . ,(lambda (ln)
+                     (cons (car ln)
+                           ((o list
+                               add-inline-markup
+                               escape-entities
+                               string-join)
+                            (cdr ln))))))
+    (default
+      (line . ,(o list
+                  add-inline-markup
+                  escape-entities
+                  string-join))
+      (stanza . ,string-join))))
@@ -36,19 +36,34 @@
          ((null? words)                            ; empty line
           (parse-lines (cdr lines) doc))
          ((equal? (car words) "```")               ; verbatim
-          (parse-verbatim (cdr lines) doc '()))
+          ;; Format for verbatim header:
+          ;; ``` ?html | command ...
+          ;;     -- only run command on block with html output.
+          ;;        other outputs process the block normally
+          ;; ``` ?!html | command ...
+          ;;     -- only run command on block when *not* outputting html.
+          ;;        html processes the block normally
+          ;; ``` ?:html | command ...
+          ;;     -- like ?html, but ignore the block in non-html outputs.
+          ;;;; FIXME: I think this necessitates a special emit-verbatim
+          ;;;; function.
+          (parse-verbatim (cdr lines) doc '()
+                   #;       (if (< 1 (length words))
+                              (cons 'verb (cdr words))
+                          'verb)
+                          'verb))
          (else                                     ; another line type
           (apply parse-stanza lines doc '() (line-type words)))))))
 
-(define (parse-verbatim lines doc block)
-  (define (close-verbatim) (cons (cons 'verb (reverse block)) doc))
+(define (parse-verbatim lines doc block bhead)
+  (define (close-verbatim) (cons (cons bhead (reverse block)) doc))
   (cond
-   ((null? lines)                                  ; end of document
+   ((null? lines)                       ; end of document
     (parse-lines lines (close-verbatim)))
-   ((equal? (car lines) "```")                     ; end of verbatim block
+   ((equal? (car lines) "```")          ; end of verbatim block
     (parse-lines (cdr lines) (close-verbatim)))
-   (else                                           ; verbatim block continues
-    (parse-verbatim (cdr lines) doc (cons (list (car lines)) block)))))
+   (else                                ; verbatim block continues
+    (parse-verbatim (cdr lines) doc (cons (list (car lines)) block) bhead))))
 
 (define (parse-stanza lines doc stanza st-type
                       #!optional (st-inlines '()) (st-words cdr))
@@ -2,6 +2,7 @@
 
   (import scheme (chicken base)
           (chicken condition)
+          (only (chicken irregex) irregex-replace/all)
           (chicken string))
 
   (define-syntax define-public
@@ -34,6 +35,15 @@
             (apply alist-walk (cdr kv) (cdr keys)))))))
 
   (define (string-join ss #!optional (sep " "))
-  (string-intersperse ss sep)))
+    (string-intersperse ss sep))
+
+  (define (flush-lines-left lines)
+    (irregex-replace/all '(: bol (* space))
+                         (string-join lines) ""))
+
+  (define (join-lines lines)
+    (apply string-append lines))
+
+  )
 
 
@@ -5,7 +5,7 @@
         (jimmy util)
         (only (chicken io) read-string)
         (only (chicken port) with-output-to-string)
-        (only (chicken string) string-translate*))
+        (only (chicken string) string-translate* string-intersperse))
 
 ;; templates are strings with variables interpolated with "{{variables}}"
 
@@ -1,80 +1,126 @@
 (import scheme
         (chicken base)
         (chicken load)
+        (chicken pathname)
         (chicken port)
         (chicken process-context)
         test)
 
+(define test-dir (or (get-environment-variable "TESTS")
+                     "tests"))
+
 ;;; Setup
 
 (import (jimmy emit)
         (jimmy read)
         #;(jimmy wrap))
 
-(define test-doc #<<end-document
-: title a test document
-: date 2024-05-13T03:02:45Z
-: uuid b3daebf1-440b-4828-a4d9-9089c7bd7c61
-
-# a test document of some kind
+(test-begin)
 
-here is a test document.
-it has paragraphs
-=> example.com with links!
-and other things.
+;;; Reading
 
-## a code example
-```
-for (a=1;a<=4;a++) {
-        printf("%d\n", a);
-}
-```
+(define test-file (make-pathname (list test-dir) "test" "gmi"))
+(define expected-doc
+  '((meta ("title" "a" "test" "document")
+          ("date" "2024-05-13T03:02:45Z")
+          ("uuid" "b3daebf1-440b-4828-a4d9-9089c7bd7c61"))
+    (hdr1 ("a" "test" "document" "of" "some" "kind"))
+    (para ("here" "is" "a" "test" "document.")
+          ("it" "has" "paragraphs")
+          (link "example.com" "with" "links!")
+          ("and" "other" "things."))
+    (hdr2 ("a" "code" "example"))
+    (verb ("for (a=1;a<=4;a++) {") ("\tprintf(\"%d\\n\", a);") ("}"))
+    (hdr3 ("other" "examples"))
+    (quot ("a" "blockquote" "is" "a" "quote") ("that" "is" "blocky."))
+    (list ("list" "1") ("list" "2") ("list" "3"))
+    (link ("example.com" "link" "list" "1")
+          ("example.com" "link" "list" "2")
+          ("example.com" "link" "list" "3"))
+    (para ("ok," "now" "for" "another" "test:")
+          ("will" "*strong*" "in-line" "text" "be" "converted?")
+          ("as" "well" "as" "`code`," "_emph_" "and" "such?")
+          ("what" "if" "*i" "_nest_" "them*")
+          ("what" "if" "*i" "_nest" "them*" "wrong_" "?")
+          ("what" "about" "*breaking" "them")
+          ("over" "two" "lines?*"))))
+(define actual-doc (with-input-from-file test-file parse))
 
-### other examples
+(test "read" expected-doc actual-doc)
 
-> a blockquote is a quote
-> that is blocky.
+(define doc expected-doc)
 
-* list 1
-* list 2
-* list 3
-=> example.com link list 1
-=> example.com link list 2
-=> example.com link list 3
+;;; Emitting
 
-ok, now for another test:
-will *strong* in-line text be converted?
-as well as `code`, _emph_ and such?
-what if *i _nest_ them*
-what if *i _nest them* wrong_ ?
-what about *breaking them
-over two lines?*
-end-document
-)
+(test-group "gemini"
+  (define expected-gmi
+    (string-append "# a test document of some kind\n\n"
+                   "here is a test document. it has paragraphs \n"
+                   "=> example.com with links!\n"
+                   "and other things.\n\n"
+                   "## a code example\n\n"
+                   "```\nfor (a=1;a<=4;a++) {\n"
+                   "\tprintf(\"%d\\n\", a);"
+                   "\n}\n```\n\n"
+                   "### other examples\n\n"
+                   "> a blockquote is a quote that is blocky.\n\n"
+                   "* list 1\n"
+                   "* list 2\n"
+                   "* list 3\n\n"
+                   "=> example.com link list 1\n"
+                   "=> example.com link list 2\n"
+                   "=> example.com link list 3\n\n"
+                   "ok, now for another test: "
+                   "will *strong* in-line text be converted? "
+                   "as well as `code`, _emph_ and such? "
+                   "what if *i _nest_ them* what if *i _nest them* wrong_ ? "
+                   "what about *breaking them over two lines?*\n\n"))
+  (define actual-gmi (with-output-to-string (lambda () (emit doc))))
+  (test "emit" expected-gmi actual-gmi))
 
-;;; Tests
+;;; HTML
 
-(test "read"
-      '((meta ("title" "a" "test" "document")
-              ("date" "2024-05-13T03:02:45Z")
-              ("uuid" "b3daebf1-440b-4828-a4d9-9089c7bd7c61"))
-        (hdr1 ("a" "test" "document" "of" "some" "kind"))
-        (para "here is a test document." "it has paragraphs" (link "example.com"
-                                                                   "with"
-                                                                   "links!")
-              "and other things.")
-        (hdr2 ("a" "code" "example"))
-        (verb "for (a=1;a<=4;a++) {" "\tprintf(\"%d\\n\", a);" "}")
-        (hdr3 ("other" "examples"))
-        (quot ("a" "blockquote" "is" "a" "quote") ("that" "is" "blocky."))
-        (list ("list" "1") ("list" "2") ("list" "3"))
-        (link ("example.com" "link" "list" "1") ("example.com" "link" "list"
-                                                 "2") ("example.com" "link"
-                                                       "list" "3"))
-        (para "ok, now for another test:" "will *strong* in-line text be
-converted?" "as well as `code`, _emph_ and such?" "what if *i _nest_ them*"
-"what if *i _nest them* wrong_ ?" "what about *breaking them" "over two
-lines?*"))
-      (call-with-input-string test-doc parse))
+(test-group "html"
+  (import (jimmy html))
+  (define expected-html
+    (string-append "<h1>a test document of some kind</h1>\n"
+                   "<p>\n"
+                   " here is a test document.\n"
+                   " it has paragraphs\n"
+                   " <a href=\"example.com\">with links!</a>\n"
+                   " and other things.\n"
+                   "</p>\n"
+                   "<h2>a code example</h2>\n"
+                   "<pre><code>for (a=1;a<=4;a++) {\n"
+                   "\tprintf(\"%d\\n\", a);\n"
+                   "}\n"
+                   "</code></pre>\n"
+                   "<h3>other examples</h3>\n"
+                   "<blockquote>\n"
+                   " a blockquote is a quote\n"
+                   " that is blocky.\n"
+                   "</blockquote>\n"
+                   "<ul>\n"
+                   " <li>list 1</li>\n"
+                   " <li>list 2</li>\n"
+                   " <li>list 3</li>\n"
+                   "</ul>\n"
+                   "<ul>\n"
+                   " <li><a href=\"example.com\">link list 1</a></li>\n"
+                   " <li><a href=\"example.com\">link list 2</a></li>\n"
+                   " <li><a href=\"example.com\">link list 3</a></li>\n"
+                   "</ul>\n"
+                   "<p>\n"
+                   " ok, now for another test:\n"
+                   " will <b>strong</b> in-line text be converted?\n"
+                   " as well as <code>code</code>, <i>emph</i> and such?\n"
+                   " what if <b>i <i>nest</i> them</b>\n"
+                   " what if <b>i <i>nest them</b> wrong</i> ?\n"
+                   " what about *breaking them\n"
+                   " over two lines?*\n"
+                   "</p>\n"))
+  (define actual-html (with-output-to-string (lambda () (emit doc))))
+  (test "emit html" expected-html actual-html))
 
+(test-end)
 (test-exit)