merge with main

README.md

@@ -164,10 +164,12 @@ legend: ⭐ priority , 🧪 tests
 If you are interested in "fontend" / desktop technologies check out separate project that works on extending Rye lanuage with GUI, Game engine and a Webview. It 
 integrates these cool Go libraries:
 
-  * Fyne - Cross platform Material design inspired GUI framework ⭐ 
+  * Fyne - Cross platform Material design inspired GUI framework ⭐⭐ 
   * Ebitengine - 2d game engine
   * Webview - Webview GUI
 
+![Fyne Feedback example](https://ryelang.org/rye-fyne-2.png)
+
 **[Visit Rye-front repo](https://github.com/refaktor/rye-front)**
 
 ## Follow development

code-of-repo-conduct.md

@@ -0,0 +1,34 @@
+# Code of repository conduct
+
+With recent HN post a lot of new people came to follow development of Rye. We probably all want Rye updates and general state of Rye repository to 
+be "a good product" to follow too. 
+
+Here we try to come up with some directives, we could follow to achieve this. Everybody is are welcome to propose changes.
+
+We shouldn't have too many principles (or rules), we are not trying to create a bureaucracy here. Just the really necessary ones, the rest is advice or knowledge base ... 
+
+## Principles we really try to follow 
+
+1. Main branch shouldn't be left in a failed state (the red cross)
+2. PRs that are meant for merging should also pass all checks
+3. Each new builtin should have a docstring written and at least few tests
+
+## Q&A
+
+### How to create commits / PR-s / merge them so that feed and release notes are as informative as possible?
+
+_work-in-progress-text-propose-changes_
+
+If there is only one commit per PR then the message is stored and used (check this).
+
+If there are multiple commits's then messages get lost unless we merge with *Squash and merge*. This joins messages together and uses them (positive) but also joins all code changes into one commit. 
+A) If reason for multiple commits is iterating on a set of changes, or making them comple (adding tests, making golint-ci pass, ...)  then this makes sense. 
+B) If PR is composed of multiple commits each for different set of changes then some information is lost with Squash.
+
+* Should the choice be per-case based?
+* What to do in case (B) then?
+* Is there a third option?
+* Should in case of (B) these be multiple PR-s anyway?
+
+
+

evaldo/builtins_spreadsheet.go

@@ -35,7 +35,6 @@ var Builtins_spreadsheet = map[string]*env.Builtin{
 						cols[i] = k.Value
 					}
 				}
-				fmt.Println(cols)
 				spr := env.NewSpreadsheet(cols)
 				switch data1 := arg1.(type) {
 				case env.Block:
@@ -78,7 +77,7 @@ var Builtins_spreadsheet = map[string]*env.Builtin{
 
 	"to-spreadsheet": {
 		Argsn: 1,
-		Doc:   "Create a spreadsheet by accepting block of dicts",
+		Doc:   "Create a spreadsheet by accepting block or list of dicts",
 		Fn: func(ps *env.ProgramState, arg0 env.Object, arg1 env.Object, arg2 env.Object, arg3 env.Object, arg4 env.Object) (res env.Object) {
 			switch block := arg0.(type) {
 			case env.Block:
@@ -118,8 +117,45 @@ var Builtins_spreadsheet = map[string]*env.Builtin{
 				}
 				return *spr
 
+			case env.List:
+				data := block.Data
+				if len(data) == 0 {
+					return MakeBuiltinError(ps, "List is empty", "to-spreadsheet")
+				}
+				k := make(map[string]struct{})
+				for _, obj := range data {
+					switch dict := obj.(type) {
+					case map[string]any:
+						for key := range dict {
+							k[key] = struct{}{}
+						}
+					default:
+						return MakeBuiltinError(ps, "List must contain only dicts", "to-spreadsheet")
+					}
+				}
+				var keys []string
+				for key := range k {
+					keys = append(keys, key)
+				}
+				spr := env.NewSpreadsheet(keys)
+				for _, obj := range data {
+					row := make([]any, len(keys))
+					switch dict := obj.(type) {
+					case map[string]any:
+						for i, key := range keys {
+							data, ok := dict[key]
+							if !ok {
+								data = env.Void{}
+							}
+							row[i] = data
+						}
+					}
+					spr.AddRow(*env.NewSpreadsheetRow(row, spr))
+				}
+				return *spr
+
 			default:
-				return MakeArgError(ps, 1, []env.Type{env.BlockType}, "to-spreadsheet")
+				return MakeArgError(ps, 1, []env.Type{env.BlockType, env.ListType}, "to-spreadsheet")
 			}
 		},
 	},