Fix issues #50, #53, and #58

Issue #50: Add validation for invalid page separators
- Add detection for en-dash, em-dash, minus sign, and other invalid dash characters in page numbers
- Provide suggested corrections by replacing invalid dashes with standard hyphens
- Return error message when invalid dash characters are detected

Issue #53: Fix capitalization validation with surrounding punctuation
- Add strip_leading_trailing_non_letters() helper function to extract leading/trailing non-alphabetic characters
- Update format_title() to strip punctuation before checking caps.txt, then reapply after adding braces
- Update format_journal_name() with same fix
- Fixes incorrect brace placement like {(BHI}) → ({BHI})

Issue #58: Add MacOS TeXShop/TeX Live setup instructions
- Add new section "MacOS Setup with TeXShop and TeX Live" to README
- Document TeX Live's ~/Library/texmf/ approach for GUI applications
- Explain why environment variables don't work with Mac GUI apps
- Provide step-by-step instructions with symbolic link setup
- Update table of contents with new subsections

Author: claude

README.md

@@ -16,6 +16,8 @@ The main bibtex file ([cdl.bib](https://raw.githubusercontent.com/ContextLab/CDL
   - [`compare`](#compare)
   - [`commit`](#commit)
 - [Using the bibtex file as a common bibliography for all *local* LaTeX files](#using-the-bibtex-file-as-a-common-bibliography-for-all-local-latex-files)
+  - [General Unix/Linux Setup (Command Line Compilation)](#general-unixlinux-setup-command-line-compilation)
+  - [MacOS Setup with TeXShop and TeX Live](#macos-setup-with-texshop-and-tex-live)
 - [Using the bibtex file on Overleaf](#using-the-bibtex-file-on-overleaf)
 - [Acknowledgements](#acknowledgements)
 
@@ -190,6 +192,8 @@ called, and a pull request must be submitted in order to integrate the changes
 into the main ContextLab fork.
 
 # Using the bibtex file as a common bibliography for all *local* LaTeX files
+
+## General Unix/Linux Setup (Command Line Compilation)
 1. Check out this repository to your home directory
 2. Add the following lines to your `~/.bash_profile` (or `~/.zshrc`, etc.):
 ```
@@ -208,6 +212,28 @@ latex filename
 pdflatex filename
 ```
 
+## MacOS Setup with TeXShop and TeX Live
+
+Mac GUI applications like TeXShop don't execute within your shell environment, which means the environment variable approach described above won't work when compiling through the TeXShop GUI. Instead, use TeX Live's built-in support for personal files:
+
+1. Check out this repository (we'll assume you cloned it to your home directory: `~/CDL-bibliography`)
+2. Create the TeX Live personal texmf directory structure for bibliography files:
+```bash
+mkdir -p ~/Library/texmf/bibtex/bib
+```
+3. Create a symbolic link from your personal texmf directory to the CDL-bibliography repository. **Important**: You must use the absolute path (not relative paths or `~`):
+```bash
+ln -s /Users/YOUR_USERNAME/CDL-bibliography/cdl.bib ~/Library/texmf/bibtex/bib/cdl.bib
+```
+Replace `YOUR_USERNAME` with your actual macOS username, or use `$HOME` instead:
+```bash
+ln -s $HOME/CDL-bibliography/cdl.bib ~/Library/texmf/bibtex/bib/cdl.bib
+```
+4. In your .tex file, use the line `\bibliography{cdl}` to generate a bibliography using the citation keys defined in cdl.bib
+5. Compile your document using TeXShop's GUI or from the command line
+
+**Note**: This approach also works for command-line compilation, so you don't need to set up the environment variables if you use this method.
+
 # Using the bibtex file on Overleaf
 You can use [git submodules](https://blog.github.com/2016-02-01-working-with-submodules/) to maintain a reference to the cdl.bib file in this repository that you can easily keep in sync with latest version.  This avoids the need to maintain a separate .bib file in each Overleaf project.
 

bibcheck/helpers.py

@@ -427,6 +427,21 @@ def roman2int(s):
 
 
 def valid_pages(p):
+    # Check for invalid dash characters (en-dash, em-dash, minus sign, etc.)
+    invalid_dashes = {
+        '\u2013': 'en-dash (–)',
+        '\u2014': 'em-dash (—)',
+        '\u2212': 'minus sign (−)',
+        '\u2010': 'hyphen (‐)',
+        '\u2011': 'non-breaking hyphen (‑)',
+    }
+
+    for dash_char, dash_name in invalid_dashes.items():
+        if dash_char in p:
+            # Return False with error message
+            suggested_fix = p.replace(dash_char, '-')
+            return False, [p, suggested_fix]
+
     valid, kind, val = valid_page(p)
     if valid: #"single" page
         return True, [p, p]
@@ -483,21 +498,25 @@ def format_journal_name(n, key=journal_key, force_caps=force_caps):
     #words = ['-'.join([format_journal_name(x) for x in w.split('-')]) if len(w.split('-')) > 1 else w for w in words] #deal with hyphens
 
     for i, w in enumerate(words):
-        words[i] = w.capitalize()
-        
-        #deal with hyphens
-        if len(w.split('-')) > 1:
-            words[i] = '-'.join(format_journal_name(c, key=key, force_caps=force_caps) for c in w.split('-'))
-        
-        if (i > 0) and (w.lower() in uncaps):
-            words[i] = words[i].lower()
-        
-        correct_caps = [f for f in force_caps if f.lower() == remove_non_letters(w.lower())]
+        # Strip leading/trailing non-letters before checking caps
+        prefix, core, suffix = strip_leading_trailing_non_letters(w)
+        correct_caps = [f for f in force_caps if f.lower() == remove_non_letters(core.lower())]
+
         if len(correct_caps) >= 1:
             c = correct_caps[-1]
             if not (c[0] == '{' and c[-1] == '}'):
-                c = insert_non_letters('{' + c + '}', remove_curlies(w, join=' '))
-            words[i] = c
+                c = insert_non_letters('{' + c + '}', remove_curlies(core, join=' '))
+            # Add back prefix and suffix
+            words[i] = prefix + c + suffix
+        else:
+            words[i] = w.capitalize()
+
+            #deal with hyphens
+            if len(w.split('-')) > 1:
+                words[i] = '-'.join(format_journal_name(c, key=key, force_caps=force_caps) for c in w.split('-'))
+
+            if (i > 0) and (w.lower() in uncaps):
+                words[i] = words[i].lower()
     return ' '.join(words)
 
 #rearrange author name (first middle last suffix)
@@ -624,7 +643,36 @@ def before_letters(s, c): #true if c occurs before the first letter in s
 def after_letters(s, c): #true if c occurs after the last letter in s
     return before_letters(s[::-1], c)
 
-def insert_non_letters(x, y):    
+def strip_leading_trailing_non_letters(s):
+    """Strip leading and trailing non-alphabetic characters from a string.
+    Returns (prefix, core, suffix) where core contains only letters and internal punctuation."""
+    if len(s) == 0:
+        return '', '', ''
+
+    # Find first letter
+    first_letter = -1
+    for i, c in enumerate(s):
+        if c.lower() in ascii_lowercase:
+            first_letter = i
+            break
+
+    if first_letter == -1:  # No letters found
+        return s, '', ''
+
+    # Find last letter
+    last_letter = -1
+    for i in range(len(s) - 1, -1, -1):
+        if s[i].lower() in ascii_lowercase:
+            last_letter = i
+            break
+
+    prefix = s[:first_letter]
+    core = s[first_letter:last_letter + 1]
+    suffix = s[last_letter + 1:]
+
+    return prefix, core, suffix
+
+def insert_non_letters(x, y):
     z = ''
     i = 0 #position in x
     j = 0 #position in y
@@ -641,15 +689,15 @@ def insert_non_letters(x, y):
             j += 1
         else: #x[i] and y[j] are both in ascii_lowercase but x[i] != x[j] -- throw an error
             raise Exception(f'"{y}" is not a compatable template for "{x}"')
-    
+
     #insert trailing punctuation from y
     if (j < len(y)) and (remove_non_letters(y[j:]) == ''):
         z += y[j:]
-    
+
     #insert trailing punctuation from x
     if (i < len(x)) and (remove_non_letters(x[i:]) == ''):
         z += x[i:]
-    
+
     return z
 
 def format_title(title):
@@ -689,15 +737,19 @@ def ends_in_punctuation(s):
 
         #leave "a" and specified caps unchanged
         if w.lower() == 'a' or (before_letters(w, '{') and after_letters(w, '}')):
-            reformatted_title.append(w)                  
+            reformatted_title.append(w)
         #if w contains curly braces, just append it unchanged
         elif (w.count('{') > 0) or (w.count('}') > 0):
-            reformatted_title.append(w)        
+            reformatted_title.append(w)
         else:
-            caps_match = [f for f in force_caps if f.lower() == remove_non_letters(w.lower())]
+            # Strip leading/trailing non-letters before checking caps
+            prefix, core, suffix = strip_leading_trailing_non_letters(w)
+            caps_match = [f for f in force_caps if f.lower() == remove_non_letters(core.lower())]
             if len(caps_match) > 0:
-                w = insert_non_letters('{' + caps_match[-1] + '}', w)
-            elif not ends_in_punctuation(prev_w):                
+                # Apply braces only to the core, then add back prefix and suffix
+                core_with_braces = insert_non_letters('{' + caps_match[-1] + '}', remove_curlies(core, join=' '))
+                w = prefix + core_with_braces + suffix
+            elif not ends_in_punctuation(prev_w):
                 w = w.lower()
             reformatted_title.append(w)
         prev_w = w