A lot of stuff together

Author: mpourmpoulis

CHANGELOG.md

@@ -10,6 +10,8 @@
 
 * PopUpErrorAction has been added in order to inform the user in  cases where to cannot parse the source code.
 
+* Sublime settings file has been added, from which you can turn on or off, showing the main result when it is not visible as well as the pop-ups with error information.
+
 ### Fixed
 
 * fixed the empty brackets bug in the repair modulebugs with ":" addressed, needs more testing and a subsequent bug
@@ -28,7 +30,11 @@
 
 * fixed a bug in the above/below argument command, which occasionally caused of by one  errors. 
 
-* in the info module, fixed a bug in the get_body function where ast.IfExp was missing for some reason. 
+* in the info module, fixed a bug in the get_body function where ast.IfExp was missing for some reason
+
+* Yet another forwarding bug in the argument module:(
+
+*  fix a bug in the big region of interest  module which caused off by one error in the above below command, when the original selection was in an empty line.
 
 ### Changed
 
@@ -60,6 +66,8 @@
 
 * In the same spirit, the query module has been modified and has a new attribute in order to store the above exception. The application module checks if this attribute HUD stored exception and displays an error through the PopUpErrorAction!
 
+* The interface module has been modified to accept settings in its constructor.
+
 ## [0.0.3] - 2019-11-18
 
 important patch

doc/SelectArgument.md

@@ -103,11 +103,28 @@ Choice("vertical_direction",{
 ```
 and ndir is an interger specifying how many lines (relative to the current) up or down your roi is.
 
+
 But why both "above" and "up"? The difference lies in that above only counts "interesting lines", that is(physical) lines containing function calls. The following example should clarify this:
 
 ![](./gif/arg5.gif)
 
-Though if we want to be more precise, we count lines that contain the beginning of function calls! this is important because logical lines can extend over multiple "physical" lines. The last example in the gif contains such an example, physical lines still determine the line to which above/below refer  and result and  alternatives from that physical line will be prioritized, but alternatives will also be offered from  the whole logical line! 
+Though if we want to be more precise, we count lines that contain the beginning of function calls! this is important because logical lines can extend over multiple "physical" lines. The last example in the gif contains such an example,but to give you a better idea:
+
+![](./gif/arg9.gif)
+
+ so you can see that
+	* physical lines still determine the line to which above/below refer  
+	* result and  alternatives from that physical line are  being prioritized, but 
+	* The whole logical line is scanned for alternatives ! 
+
+In a more complex scenario, if there are multiple functional calls in the physical line and you up for using an ordinal adjective, in order to to preserve sanity, the main result will be decided by taking into consideration the order of lexical appearance:
+
+![](./gif/arg10.gif)
+
+
+Finally, even though the argument query was originally designed to operate on a single logical line, make things more consistent, an exception was made for the case when the physical line that you are targeting contains more than one logical lines. In such a case, all the logical lines are processed in the monitor similar to the previous case, with important distinction that both the result and the alternatives must come from this physical line, where as in the previous case either one of them can come from other physical lines, if no suitable candidate is found.
+
+![](./gif/arg11.gif)
 
 
 

doc/SelectBigROI.md

@@ -177,6 +177,14 @@ Sub indexing functionality has been expanded to include picking up parts of stri
 
 ![](./gif/big10.gif)  
 
+We can pick up parts from the URL, individual words or letters, or part of a camel or snake case. this feature is still immatur  and needs more work, but I am planning to improve and also expand it with the ability to select a whole range.
+
+Also something that was kind of missing,you can now select a subset of an arithmetic expression :
+
+![](./gif/big12.gif)  
+
+Once again you need to pay attention to operator precedence and as you can see there are some edge cases that need to be fixed.
+
 Finally, we clarify one more thing! What about relative vertical offsets when using above? We know that these abstract vertical keywords only count interesting lines, but what do we count as interesting here? To stay compatible with all of the above, we count all lines containing our desired big region of interest regardless of whether we can extract or not from them information with the sub index! As an example:
 
 ![](./gif/big9.gif)  

doc/gif/big12.gif

@@ -13,15 +13,17 @@ class SelectionAction(InterfaceAction):
 	def __init__(self, region):
 		self.data={"region":region}
 
-	def execute(self,view,sublime,**kwargs):
+	def execute(self,view,settings,sublime,**kwargs):
 		region =  self.data["region"]
 		if region:
 			view.sel().clear()
 			if not isinstance(region,list):
 				region = [region]
 			for r in region:
 				view.sel().add(sublime.Region(r[0],r[1]))
-			view.show(sublime.Region(region[0][0],region[0][1]))
+			if settings.get("show_visible",False):
+				view.show(sublime.Region(region[0][0],region[0][1]))
+			
 
 
 
@@ -213,11 +215,13 @@ class PopUpErrorAction(InterfaceAction):
 	"""docstring for DisplayErrorAction"""
 	def __init__(self, text):
 		self.text = text
-	def execute(self,view, sublime,**kwargs):
+	def execute(self,view,settings, sublime,**kwargs):
+		if not settings.get("show_error",False):
+			return 
 		final_text = "<p></p><h>Something is off!</h>" + "<p>" + html.escape(self.text) + "</p>"
 		print(" inside final text processing ",final_text)
 		def on_hide():
-			view.show_popup(final_text,max_width=1024, max_height=10000, flags= sublime.COOPERATE_WITH_AUTO_COMPLETE)
+			view.show_popup(final_text,max_width=1024, max_height=10000, flags= sublime.HIDE_ON_MOUSE_MOVE_AWAY)
 		view.show_popup(final_text,max_width=1024, max_height=10000, 
 			flags= sublime.COOPERATE_WITH_AUTO_COMPLETE,on_hide = on_hide)
 		print(view.is_popup_visible())

interface/common/actions.py

@@ -4,12 +4,13 @@
 
 class Interface():
 	"""docstring for Interface"""
-	def __init__(self,view,window,edit,sublime):
+	def __init__(self,view,window,edit,sublime,settings):
 		self.view = view
 		self.window = window
 		self.edit = edit
 		self.sublime = sublime
 		self.actions = []
+		self.settings = settings
 
 	def get_view_information(self):
 		return ViewInformation(self.view,self.sublime)
@@ -24,7 +25,7 @@ def respond_to_query(self,query_description):
 		application = Application.get_application(self.view.id())
 		application.respond_to_query(self,query_description)
 		parameters = {
-			"view":self.view,"window":self.window,"edit":self.edit,"sublime":self.sublime
+			"view":self.view,"window":self.window,"edit":self.edit,"sublime":self.sublime,"settings":self.settings,
 		}
 		for action in self.actions:
 			action.execute(**parameters)

interface/interface.py

@@ -294,7 +294,7 @@ def split_string(s,even_letters = True):
 	second_attempt = [x  for x in re.split("[_]",s) if not x.isspace()]
 	if len(second_attempt) > 1:
 		return second_attempt
-# https://stackoverflow.com/questions/29916065/how-to-do-camelcase-split-in-python Jossef Harush 
+	# https://stackoverflow.com/questions/29916065/how-to-do-camelcase-split-in-python answer from Jossef Harush 
 	third_attempt = re.sub('([A-Z][a-z]+)', r' \1', re.sub('([A-Z]+)', r' \1', s)).split()
 	if len(third_attempt) > 1:
 		return third_attempt

library/info.py

@@ -146,8 +146,6 @@ def get_timestamp(self):
 		return len(self.history)
 
 	def forward(self,value,start_time = 0,end_time = 0):
-		print(" value is", value,"over")
-		print(value)
 		if value is None: 
 			return None
 		end_time = end_time if end_time!=0 else len(self.history)	

library/modification.py

@@ -56,7 +56,7 @@ def partially_parse(code, m = None, atok = None,rethrow_exception = False):
 			print(" go to the field\n",m.current_code)
 			print(" error was\n",str(e),"\n",e)
 			if rethrow_exception :
-				raise e_first
+				raise e
 			return None,None,None,None
 
 ###############################
@@ -66,15 +66,13 @@ def line_partial(code,offset):
 	atok = asttokens.ASTTokens(parse=False, source_text=code)
 	origin = atok.get_token_from_offset(offset)
 	left, right = expand_to_line_or_statement(atok,origin)
-	print( left, right )
 	m = ModificationHandler(code)
 	m.modify_from(0,(0, left.startpos),"")
 	if right.string == ":":
 			m.modify_from(0,(right.endpos,len(code)+1),"pass")
 	else:
 			m.modify_from(0,(right.endpos,len(code)+1),"")
 	m.update()
-	print("history  is",m.history)
 	return partially_parse(m.current_code,m)
 
 ################################################################################################

library/partial.py

@@ -4,6 +4,7 @@
 import sublime
 import sublime_plugin
 
+settings = {}
 # making sure our dependencies are in the path
 sys.path.insert(0,os.path.join(os.path.dirname(__file__), 'third_party'))
 
@@ -12,16 +13,25 @@
 # removing our dependencies from the path so as not to interfere with other packages
 sys.path.remove(os.path.join(os.path.dirname(__file__), 'third_party'))
 
+
+def plugin_loaded():
+	global settings
+	settings = sublime.load_settings("python_voice_coding_plugin.sublime-settings")
+
+	
+
 class PythonVoiceCodingPluginCommand(sublime_plugin.TextCommand):
 	def run(self, edit,arg):
 		self.action_one(edit,arg)
 
 	def action_one(self, edit,arg):
+		global settings
 		interface = Interface(
 			sublime = sublime,
 			view = self.view,
 			window = sublime.active_window(),
-			edit = edit
+			edit = edit,
+			settings = settings
 		)
 		interface.respond_to_query(arg)