close #2 import duplicates are detected and ignored

ReadMe.md

@@ -2,15 +2,16 @@
 CLI tracker to prioritize your limited time by ELO ranking media.
 
 ## Initial Ideas
-- [ ] Mark books read/unread/in-progress.
+- [x] Mark books read/unread/in-progress.
 - [x] Show top 10 / bottom 10. Show only in-progress or unread.
 - [ ] Rank only in-progress or unread.
 - [x] Track books by title, author, and series. Default is to recognize "by" or "(series)" and automatically sort into relevant fields.
   - [x] But also almost nothing is actually *needed*.
-- [x] Store progress as a numerical value? (Unread = 0, read = 1, in-progress in-between? Too easy to forget how it works?)
 - [x] Store pages as an option,
-  - [ ] allow a weighted priority based on remaining percentage or whatever.
+  - [x] allow a weighted priority based on remaining percentage or whatever.
 - [x] Store items as an array instead of object, because that makes the JSON output more consistent -> smaller JSON commit differences.
+      - [ ] I am not sure this actually helps much when EACH ITEM is still an
+            object. I need to make the JSON export sort its keys.
 - [ ] Re-implement ELO ranking.
 - [ ] Display filter/sort options to output lists. (Focus only on top 10 items?)
 
@@ -22,6 +23,12 @@ I accept pull requests and issues, but management of the code is handled there.
 
 Licensing? I don't care. Do whatever you want. IP rights are stupid.
 
+## Ideas
+- [ ] When importing, allow user to make a choice when duplicates are detected
+      instead of rejecting them automatically.
+- [ ] When importing, a duplicate should still be checked for missing data, and
+      merge rather than ignore.
+
 ## Book Data Structure
 ```json
 {

asymptome.lua

@@ -249,14 +249,35 @@ local update_book = function(book)
   return update_progress_prompt(book)
 end
 
+local book_exists = function(data, book)
+  for i = 1, #data.books do
+    local current = data.books[i]
+    if not (book.title == current.title) then
+      break
+    end
+    if not (book.author == current.author) then
+      break
+    end
+    if not (book.series == current.series) then
+      break
+    end
+    if not (book.genre == current.genre) then
+      break
+    end
+    return current -- we return the object instead of true so it can be updated if desired
+  end
+  return false
+end
+
 local import_json = function(data, file_name)
   import = load_json(file_name)
   if import.books then
 
     for i = 1, #import.books do
-      data.books[#data.books + 1] = import.books[i]
+      -- TODO allow updating metadata instead of just ignoring
-      -- TODO detect and ignore duplicates (would be better to provide user an option for what to do with duplicates)
+      if not book_exists(data, import.books[i]) then
-      --  default should accept highest progress of duplicates?
+        data.books[#data.books + 1] = import.books[i]
+      end
     end
 
   else -- assume we are importing from decider.lua's format
@@ -292,8 +313,10 @@ local import_json = function(data, file_name)
           end
         end
 
-        -- TODO detect duplicates before adding them
+        -- TODO allow updating metadata (this and the above need to be split into its own function)
-        data.books[#data.books + 1] = book
+        if not book_exists(data, book) then
+          data.books[#data.books + 1] = book
+        end
       end
       run()
     end