aboutsummaryrefslogtreecommitdiff
path: root/modules/cj-org-text-lib.el
diff options
context:
space:
mode:
authorCraig Jennings <c@cjennings.net>2026-05-10 15:25:27 -0500
committerCraig Jennings <c@cjennings.net>2026-05-10 15:25:27 -0500
commit0f9e308745b77ca589e5b25f415121058b9f08ee (patch)
tree3080ab24298b6b40b0542acce185312b906c2177 /modules/cj-org-text-lib.el
parenta0e5edfb3d77f34d0f71bdc9bb8bc3a563ee6436 (diff)
downloaddotemacs-0f9e308745b77ca589e5b25f415121058b9f08ee.tar.gz
dotemacs-0f9e308745b77ca589e5b25f415121058b9f08ee.zip
refactor(cj-org-text): rename to cj-org-text-lib for naming consistency
Same naming-convention fix as the cj-cache rename. Org-safe text sanitizers extracted in Phase 3 went into modules/cj-org-text.el, should have followed the established `-lib' suffix convention. Rename modules/cj-org-text.el -> modules/cj-org-text-lib.el; update provide form, file header, and the two (require 'cj-org-text) call sites in calendar-sync and test-cj-org-text-sanitize. No behavior change.
Diffstat (limited to 'modules/cj-org-text-lib.el')
-rw-r--r--modules/cj-org-text-lib.el58
1 files changed, 58 insertions, 0 deletions
diff --git a/modules/cj-org-text-lib.el b/modules/cj-org-text-lib.el
new file mode 100644
index 00000000..45b9ec2e
--- /dev/null
+++ b/modules/cj-org-text-lib.el
@@ -0,0 +1,58 @@
+;;; cj-org-text-lib.el --- Pure helpers for sanitizing external text into Org -*- lexical-binding: t; -*-
+
+;; Author: Craig Jennings <c@cjennings.net>
+
+;;; Commentary:
+
+;; Pure string helpers for safely composing Org-mode content from
+;; external text (calendar event bodies, web-clipped HTML, mail
+;; subject lines, AI conversation transcripts, etc.).
+;;
+;; The shared concern is that text from outside sources can contain
+;; characters that disturb Org structure if pasted verbatim:
+;;
+;; - leading `*' creates an unintended heading,
+;; - newlines inside a property value spawn extra drawer lines,
+;; - newlines inside a heading split it into two outline entries.
+;;
+;; These helpers neutralize each pattern with predictable, testable
+;; replacements. They are pure (string in, string out, nil-safe) and
+;; have no Org-mode dependency, so they remain useful in batch and in
+;; tests without loading Org.
+
+;;; Code:
+
+(defun cj/org-sanitize-body-text (text)
+ "Sanitize TEXT for safe inclusion as Org body content.
+Replaces leading asterisks with dashes so external lines aren't
+parsed as Org headings. Handles multiple levels (`**' becomes `--').
+Returns nil for nil input."
+ (when text
+ (replace-regexp-in-string
+ "^\\(\\*+\\) "
+ (lambda (match)
+ (concat (make-string (length (match-string 1 match)) ?-) " "))
+ text)))
+
+(defun cj/org-sanitize-property-value (text)
+ "Sanitize TEXT for safe inclusion as a single Org property value.
+Collapses whitespace and newlines into single spaces and trims, so the
+result fits on one line of an Org property drawer. Returns nil for
+nil input."
+ (when text
+ (string-trim
+ (replace-regexp-in-string
+ "[[:space:]\n\r]+"
+ " "
+ text))))
+
+(defun cj/org-sanitize-heading (text)
+ "Sanitize TEXT for safe inclusion as a single Org heading title.
+Composes `cj/org-sanitize-body-text' (neutralizes leading stars) and
+`cj/org-sanitize-property-value' (flattens to a single line). Returns
+nil for nil input."
+ (cj/org-sanitize-property-value
+ (cj/org-sanitize-body-text text)))
+
+(provide 'cj-org-text-lib)
+;;; cj-org-text-lib.el ends here