What is the accessibility of a package's `Private` context variables? Planned maintenance scheduled April 17/18, 2019 at 00:00UTC (8:00pm US/Eastern) Announcing the arrival of Valued Associate #679: Cesar Manara Unicorn Meta Zoo #1: Why another podcast?How symbol lookup actually worksWhat are recommended guidelines for developing packages?How to properly handle mutual imports of multiple packages?How can Private functions be made completely opaque?Does one need to be careful about loading multiple (many) contexts or packages in the same session?WebServices context problemHow to pass rules to packagesA question regarding shadowed symbolsIs there any harm or benefit to Removing unneeded private symbols in packages?Information (??) of function defined in Package return the function with long name of variablesHow to resolve a context shadow problem (revised)
Two different pronunciation of "понял"
How to politely respond to generic emails requesting a PhD/job in my lab? Without wasting too much time
How is simplicity better than precision and clarity in prose?
Writing Thesis: Copying from published papers
How do I automatically answer y in bash script?
Is there a documented rationale why the House Ways and Means chairman can demand tax info?
How can I make names more distinctive without making them longer?
Why don't the Weasley twins use magic outside of school if the Trace can only find the location of spells cast?
If I can make up priors, why can't I make up posteriors?
Stars Make Stars
When communicating altitude with a '9' in it, should it be pronounced "nine hundred" or "niner hundred"?
Keep going mode for require-package
Using "nakedly" instead of "with nothing on"
Stop battery usage [Ubuntu 18]
Are my PIs rude or am I just being too sensitive?
What did Darwin mean by 'squib' here?
Why does this iterative way of solving of equation work?
What's the difference between (size_t)-1 and ~0?
Area of a 2D convex hull
Why is there no army of Iron-Mans in the MCU?
Notation for two qubit composite product state
How should I respond to a player wanting to catch a sword between their hands?
Slither Like a Snake
3 doors, three guards, one stone
What is the accessibility of a package's `Private` context variables?
Planned maintenance scheduled April 17/18, 2019 at 00:00UTC (8:00pm US/Eastern)
Announcing the arrival of Valued Associate #679: Cesar Manara
Unicorn Meta Zoo #1: Why another podcast?How symbol lookup actually worksWhat are recommended guidelines for developing packages?How to properly handle mutual imports of multiple packages?How can Private functions be made completely opaque?Does one need to be careful about loading multiple (many) contexts or packages in the same session?WebServices context problemHow to pass rules to packagesA question regarding shadowed symbolsIs there any harm or benefit to Removing unneeded private symbols in packages?Information (??) of function defined in Package return the function with long name of variablesHow to resolve a context shadow problem (revised)
$begingroup$
I've been reading up on how Mathematica handles contexts, $Context
, $ContextPath
, and a few of the tutorials they have on Packages.
What I'm wondering about is how the functions defined in, say, CustomPackage`
are able to access the variables in CustomPackage`Private`
.
For example,
BeginPackage["CustomPackage`"]
MyFunction::usage = "MyFunction[arg1] adds 5 to arg1."
Begin["`Private`"]
abc=5;
MyFunction[arg1_] := arg1 + abc;
End[]
EndPackage[]
When I load the package <<CustomPackage`
the $ContextPath
will have CustomPackage`
on it, but not CustomPackage`Private`
So how does MyFunction
know the value of abc
at the delayed function call (when it is called) if the Private`
context isn't on the $ContextPath
packages core-language scoping contexts
$endgroup$
add a comment |
$begingroup$
I've been reading up on how Mathematica handles contexts, $Context
, $ContextPath
, and a few of the tutorials they have on Packages.
What I'm wondering about is how the functions defined in, say, CustomPackage`
are able to access the variables in CustomPackage`Private`
.
For example,
BeginPackage["CustomPackage`"]
MyFunction::usage = "MyFunction[arg1] adds 5 to arg1."
Begin["`Private`"]
abc=5;
MyFunction[arg1_] := arg1 + abc;
End[]
EndPackage[]
When I load the package <<CustomPackage`
the $ContextPath
will have CustomPackage`
on it, but not CustomPackage`Private`
So how does MyFunction
know the value of abc
at the delayed function call (when it is called) if the Private`
context isn't on the $ContextPath
packages core-language scoping contexts
$endgroup$
add a comment |
$begingroup$
I've been reading up on how Mathematica handles contexts, $Context
, $ContextPath
, and a few of the tutorials they have on Packages.
What I'm wondering about is how the functions defined in, say, CustomPackage`
are able to access the variables in CustomPackage`Private`
.
For example,
BeginPackage["CustomPackage`"]
MyFunction::usage = "MyFunction[arg1] adds 5 to arg1."
Begin["`Private`"]
abc=5;
MyFunction[arg1_] := arg1 + abc;
End[]
EndPackage[]
When I load the package <<CustomPackage`
the $ContextPath
will have CustomPackage`
on it, but not CustomPackage`Private`
So how does MyFunction
know the value of abc
at the delayed function call (when it is called) if the Private`
context isn't on the $ContextPath
packages core-language scoping contexts
$endgroup$
I've been reading up on how Mathematica handles contexts, $Context
, $ContextPath
, and a few of the tutorials they have on Packages.
What I'm wondering about is how the functions defined in, say, CustomPackage`
are able to access the variables in CustomPackage`Private`
.
For example,
BeginPackage["CustomPackage`"]
MyFunction::usage = "MyFunction[arg1] adds 5 to arg1."
Begin["`Private`"]
abc=5;
MyFunction[arg1_] := arg1 + abc;
End[]
EndPackage[]
When I load the package <<CustomPackage`
the $ContextPath
will have CustomPackage`
on it, but not CustomPackage`Private`
So how does MyFunction
know the value of abc
at the delayed function call (when it is called) if the Private`
context isn't on the $ContextPath
packages core-language scoping contexts
packages core-language scoping contexts
edited Apr 10 at 22:10
m_goldberg
88.7k873200
88.7k873200
asked Apr 10 at 19:03
w1resw1res
23015
23015
add a comment |
add a comment |
4 Answers
4
active
oldest
votes
$begingroup$
So how does
MyFunction
know the value ofabc
at the delayed function call (when it is called) if thePrivate`
context isn't on the$ContextPath
There is a misunderstanding here. You are assuming that abc
is searched for in some context only when MyFunction[something]
is evaluated. This is not the case.
$Context
and $ContextPath
only affect how source code is parsed (not how expressions are evaluated). In other words, they only affect how the text you write in the package file is interpreted and converted into in-memory expressions. Once the package has been loaded with Get
, this interpretation has already happened. MyFunction
has been interpreted as the symbol CustomPackage`MyFunction
and abc
has been interpreted as CustomPackage`Private`abc
, according to the value of $Context
and $ContextPath
at the time each was read. These are the full names of these symbols and this is how they exist in memory.
Load the package and try this:
Block[$ContextPath,
Print@Definition[MyFunction]
]
You'll see the following printed:
CustomPackage`MyFunction[CustomPackage`Private`arg1_] :=
CustomPackage`Private`arg1+CustomPackage`Private`abc
As you can see, a context is always associated with every symbol.
$endgroup$
add a comment |
$begingroup$
Begin["`Private`"];
sets the current $Context
to "CustomPackage `Private`"
. This causes two things:
The symbol
abc
will be searched in the current context first, thus in"CustomPackage`Private`"
. Only if it is not found there, the search goes on along$ContextPath
.If no matching symbol is found this way, a new symbol
abc
is created, namely in the current$Context
which is"CustomPackage`Private`"
. So the full symbol name is"CustomPackage`Private`abc"
.
For example, running your code in a fresh kernel and executing
??MyFunction
reveals that the full definition of MyFunction
is
MyFunction[CustomPackage`Private`arg1_]:=CustomPackage`Private`arg1+CustomPackage`Private`abc
Moreover, with
?*`abc
you see that the only symbol in all contexts that matches abc
is CustomPackage`Private`abc
and has the value 5
assigned to it.
$endgroup$
add a comment |
$begingroup$
All symbols are created at load time, so when you do:
BeginPackage["X`"];
x::usage="Declaring x as an exported symbol in the X` context";
Begin["`SomePrivateContext`"];
x[a_]:=b
End[];
EndPackage[];
x
was created as X`x
but the DownValues
of x
reference X`SomePrivateContext`a
and X`SomePrivateContext`b
which were created at the time the function was defined. These symbols are unique, so that reference only ever points that a single object.
$endgroup$
add a comment |
$begingroup$
So how does MyFunction know the value of abc
at the delayed function call (when it is called)
if the Private` context isn't on the $ContextPath?
because "CustomPackage`Private`"
is the value of $Context
when MyFunction
is defined (i.e. it is not just $ContextPath
that determines what a function sees but also what is on $Context
).
TL:DR
This is a timely question because it indirectly touches upon the competing imperatives of developers and end-users. To the question itself:
The whole point of packages is that they are a form of encapsulation that allows developers to, without interferance, implement functionality for end-users without bothering them with the underlying details. In particular, the encapsulation involves controlling namespaces so that the underlying details can involve symbols that help implement the functionality but ultimately don't end-up polluting a user's namespace. All symbols defined in a "*`Private`"
namespace are created for exactly this purpose.
Hence in the OP's example, the variable abc
is an underlying detail for the implementation of the public MyFunction
. The developer needs the "detail" of abc
but this particular symbol is of no direct interest to an end-user who typically just ends up calling MyFunction[]
.
The package layout achieves this encapsulation by manipulating $ContextPath
and $Context
as the control-flow passes through the package when it is first loaded. This is described in the other answers and documentation but it can be useful to see it directly:
loc[n_] := Sow[<|
"Location" -> n,
"$Context" -> $Context,
"$ContextPath" -> $ContextPath|>];
Reap[
loc@1;
BeginPackage["CustomPackage`"];
loc@2;
MyFunction::usage = "MyFunction[arg1] adds 5 to arg1.";
Begin["`Private`"];
loc@3;
abc = 5;
MyFunction[arg1_] := arg1 + abc;
End[];
loc@4;
EndPackage[];
loc@5
]// Last // Dataset
When I load the package <the $ContextPath will have CustomPackage on it, but not CustomPackage
Private
Yes, this implements both the public exporting of all CustomPackage
functions but without polluting end-users namespaces with implementation details. In code around Location 3, all packages are cleared out thereby eliminating possible conflicts with existing abc
definitions in currently loaded packages. This is encapsulation benefitting developers but the encapsulation benefitting end-users, as observed, is that on exiting (at Location 5) $ContextPath
contains "CustomPackage`"
(to provide access to MyFunction
) but not "CustomPackage`Private`"
thereby shielding users from symbols used in MyFunction
's implementation.
A programmatic confirmation at Location 5 gives:
MemberQ["CustomPackage`"]@$ContextPath,
MemberQ["CustomPackage`Private`"]@$ContextPath,
Context["abc"]
True, False, "Global`"
At Location 3 in the control-flow, the symbol abc
is not contained in any of the contexts defined in $ContextPath
, ("CustomPackage`"
, or "System`"
) nor is it (yet) in the context defined in $Context
("CustomPackage`Private`"
). Consequently, the name abc
gets created in the context currently set to $Context
. At this location $Context
has value "CustomPackage`Private`"
and hence the symbol CustomPackage`Private`abc
is created. When the control flow then moves on to MyFunction[]
, "CustomPackage`Private`"
is still the value of $Context
so this function "sees" abc
(hence it not just $ContextPath
that determines what a function sees but what is on both$ContextPath
and $Context
).
Note how the convention of placing usage definitions at Location 2 is ostensibly for documentation purposes but its more important role is to ensure that the function goes into the package's context (see $Context
at Location 2) before subsequently being made available in the implementation and for end-users (see $ContextPath
at Locations 3 and 5).
IMO it is kind of cool how these placement protocols just work intuitively without necessarily keeping front-of-mind all the control-flow manipulations, variable-creation mechanisms etc taking place behind the scenes. Hence this means being very careful changing the framework but also IMHO the time is ripe for such extensions given that the line between users/developers may well be in the process of blurring.
$endgroup$
add a comment |
Your Answer
StackExchange.ready(function()
var channelOptions =
tags: "".split(" "),
id: "387"
;
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function()
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled)
StackExchange.using("snippets", function()
createEditor();
);
else
createEditor();
);
function createEditor()
StackExchange.prepareEditor(
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: false,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: null,
bindNavPrevention: true,
postfix: "",
imageUploader:
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
,
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
);
);
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmathematica.stackexchange.com%2fquestions%2f194963%2fwhat-is-the-accessibility-of-a-packages-private-context-variables%23new-answer', 'question_page');
);
Post as a guest
Required, but never shown
4 Answers
4
active
oldest
votes
4 Answers
4
active
oldest
votes
active
oldest
votes
active
oldest
votes
$begingroup$
So how does
MyFunction
know the value ofabc
at the delayed function call (when it is called) if thePrivate`
context isn't on the$ContextPath
There is a misunderstanding here. You are assuming that abc
is searched for in some context only when MyFunction[something]
is evaluated. This is not the case.
$Context
and $ContextPath
only affect how source code is parsed (not how expressions are evaluated). In other words, they only affect how the text you write in the package file is interpreted and converted into in-memory expressions. Once the package has been loaded with Get
, this interpretation has already happened. MyFunction
has been interpreted as the symbol CustomPackage`MyFunction
and abc
has been interpreted as CustomPackage`Private`abc
, according to the value of $Context
and $ContextPath
at the time each was read. These are the full names of these symbols and this is how they exist in memory.
Load the package and try this:
Block[$ContextPath,
Print@Definition[MyFunction]
]
You'll see the following printed:
CustomPackage`MyFunction[CustomPackage`Private`arg1_] :=
CustomPackage`Private`arg1+CustomPackage`Private`abc
As you can see, a context is always associated with every symbol.
$endgroup$
add a comment |
$begingroup$
So how does
MyFunction
know the value ofabc
at the delayed function call (when it is called) if thePrivate`
context isn't on the$ContextPath
There is a misunderstanding here. You are assuming that abc
is searched for in some context only when MyFunction[something]
is evaluated. This is not the case.
$Context
and $ContextPath
only affect how source code is parsed (not how expressions are evaluated). In other words, they only affect how the text you write in the package file is interpreted and converted into in-memory expressions. Once the package has been loaded with Get
, this interpretation has already happened. MyFunction
has been interpreted as the symbol CustomPackage`MyFunction
and abc
has been interpreted as CustomPackage`Private`abc
, according to the value of $Context
and $ContextPath
at the time each was read. These are the full names of these symbols and this is how they exist in memory.
Load the package and try this:
Block[$ContextPath,
Print@Definition[MyFunction]
]
You'll see the following printed:
CustomPackage`MyFunction[CustomPackage`Private`arg1_] :=
CustomPackage`Private`arg1+CustomPackage`Private`abc
As you can see, a context is always associated with every symbol.
$endgroup$
add a comment |
$begingroup$
So how does
MyFunction
know the value ofabc
at the delayed function call (when it is called) if thePrivate`
context isn't on the$ContextPath
There is a misunderstanding here. You are assuming that abc
is searched for in some context only when MyFunction[something]
is evaluated. This is not the case.
$Context
and $ContextPath
only affect how source code is parsed (not how expressions are evaluated). In other words, they only affect how the text you write in the package file is interpreted and converted into in-memory expressions. Once the package has been loaded with Get
, this interpretation has already happened. MyFunction
has been interpreted as the symbol CustomPackage`MyFunction
and abc
has been interpreted as CustomPackage`Private`abc
, according to the value of $Context
and $ContextPath
at the time each was read. These are the full names of these symbols and this is how they exist in memory.
Load the package and try this:
Block[$ContextPath,
Print@Definition[MyFunction]
]
You'll see the following printed:
CustomPackage`MyFunction[CustomPackage`Private`arg1_] :=
CustomPackage`Private`arg1+CustomPackage`Private`abc
As you can see, a context is always associated with every symbol.
$endgroup$
So how does
MyFunction
know the value ofabc
at the delayed function call (when it is called) if thePrivate`
context isn't on the$ContextPath
There is a misunderstanding here. You are assuming that abc
is searched for in some context only when MyFunction[something]
is evaluated. This is not the case.
$Context
and $ContextPath
only affect how source code is parsed (not how expressions are evaluated). In other words, they only affect how the text you write in the package file is interpreted and converted into in-memory expressions. Once the package has been loaded with Get
, this interpretation has already happened. MyFunction
has been interpreted as the symbol CustomPackage`MyFunction
and abc
has been interpreted as CustomPackage`Private`abc
, according to the value of $Context
and $ContextPath
at the time each was read. These are the full names of these symbols and this is how they exist in memory.
Load the package and try this:
Block[$ContextPath,
Print@Definition[MyFunction]
]
You'll see the following printed:
CustomPackage`MyFunction[CustomPackage`Private`arg1_] :=
CustomPackage`Private`arg1+CustomPackage`Private`abc
As you can see, a context is always associated with every symbol.
edited Apr 10 at 20:07
answered Apr 10 at 19:47
SzabolcsSzabolcs
164k14448949
164k14448949
add a comment |
add a comment |
$begingroup$
Begin["`Private`"];
sets the current $Context
to "CustomPackage `Private`"
. This causes two things:
The symbol
abc
will be searched in the current context first, thus in"CustomPackage`Private`"
. Only if it is not found there, the search goes on along$ContextPath
.If no matching symbol is found this way, a new symbol
abc
is created, namely in the current$Context
which is"CustomPackage`Private`"
. So the full symbol name is"CustomPackage`Private`abc"
.
For example, running your code in a fresh kernel and executing
??MyFunction
reveals that the full definition of MyFunction
is
MyFunction[CustomPackage`Private`arg1_]:=CustomPackage`Private`arg1+CustomPackage`Private`abc
Moreover, with
?*`abc
you see that the only symbol in all contexts that matches abc
is CustomPackage`Private`abc
and has the value 5
assigned to it.
$endgroup$
add a comment |
$begingroup$
Begin["`Private`"];
sets the current $Context
to "CustomPackage `Private`"
. This causes two things:
The symbol
abc
will be searched in the current context first, thus in"CustomPackage`Private`"
. Only if it is not found there, the search goes on along$ContextPath
.If no matching symbol is found this way, a new symbol
abc
is created, namely in the current$Context
which is"CustomPackage`Private`"
. So the full symbol name is"CustomPackage`Private`abc"
.
For example, running your code in a fresh kernel and executing
??MyFunction
reveals that the full definition of MyFunction
is
MyFunction[CustomPackage`Private`arg1_]:=CustomPackage`Private`arg1+CustomPackage`Private`abc
Moreover, with
?*`abc
you see that the only symbol in all contexts that matches abc
is CustomPackage`Private`abc
and has the value 5
assigned to it.
$endgroup$
add a comment |
$begingroup$
Begin["`Private`"];
sets the current $Context
to "CustomPackage `Private`"
. This causes two things:
The symbol
abc
will be searched in the current context first, thus in"CustomPackage`Private`"
. Only if it is not found there, the search goes on along$ContextPath
.If no matching symbol is found this way, a new symbol
abc
is created, namely in the current$Context
which is"CustomPackage`Private`"
. So the full symbol name is"CustomPackage`Private`abc"
.
For example, running your code in a fresh kernel and executing
??MyFunction
reveals that the full definition of MyFunction
is
MyFunction[CustomPackage`Private`arg1_]:=CustomPackage`Private`arg1+CustomPackage`Private`abc
Moreover, with
?*`abc
you see that the only symbol in all contexts that matches abc
is CustomPackage`Private`abc
and has the value 5
assigned to it.
$endgroup$
Begin["`Private`"];
sets the current $Context
to "CustomPackage `Private`"
. This causes two things:
The symbol
abc
will be searched in the current context first, thus in"CustomPackage`Private`"
. Only if it is not found there, the search goes on along$ContextPath
.If no matching symbol is found this way, a new symbol
abc
is created, namely in the current$Context
which is"CustomPackage`Private`"
. So the full symbol name is"CustomPackage`Private`abc"
.
For example, running your code in a fresh kernel and executing
??MyFunction
reveals that the full definition of MyFunction
is
MyFunction[CustomPackage`Private`arg1_]:=CustomPackage`Private`arg1+CustomPackage`Private`abc
Moreover, with
?*`abc
you see that the only symbol in all contexts that matches abc
is CustomPackage`Private`abc
and has the value 5
assigned to it.
edited Apr 10 at 19:35
answered Apr 10 at 19:25
Henrik SchumacherHenrik Schumacher
60.2k582169
60.2k582169
add a comment |
add a comment |
$begingroup$
All symbols are created at load time, so when you do:
BeginPackage["X`"];
x::usage="Declaring x as an exported symbol in the X` context";
Begin["`SomePrivateContext`"];
x[a_]:=b
End[];
EndPackage[];
x
was created as X`x
but the DownValues
of x
reference X`SomePrivateContext`a
and X`SomePrivateContext`b
which were created at the time the function was defined. These symbols are unique, so that reference only ever points that a single object.
$endgroup$
add a comment |
$begingroup$
All symbols are created at load time, so when you do:
BeginPackage["X`"];
x::usage="Declaring x as an exported symbol in the X` context";
Begin["`SomePrivateContext`"];
x[a_]:=b
End[];
EndPackage[];
x
was created as X`x
but the DownValues
of x
reference X`SomePrivateContext`a
and X`SomePrivateContext`b
which were created at the time the function was defined. These symbols are unique, so that reference only ever points that a single object.
$endgroup$
add a comment |
$begingroup$
All symbols are created at load time, so when you do:
BeginPackage["X`"];
x::usage="Declaring x as an exported symbol in the X` context";
Begin["`SomePrivateContext`"];
x[a_]:=b
End[];
EndPackage[];
x
was created as X`x
but the DownValues
of x
reference X`SomePrivateContext`a
and X`SomePrivateContext`b
which were created at the time the function was defined. These symbols are unique, so that reference only ever points that a single object.
$endgroup$
All symbols are created at load time, so when you do:
BeginPackage["X`"];
x::usage="Declaring x as an exported symbol in the X` context";
Begin["`SomePrivateContext`"];
x[a_]:=b
End[];
EndPackage[];
x
was created as X`x
but the DownValues
of x
reference X`SomePrivateContext`a
and X`SomePrivateContext`b
which were created at the time the function was defined. These symbols are unique, so that reference only ever points that a single object.
answered Apr 10 at 19:20
b3m2a1b3m2a1
28.8k359166
28.8k359166
add a comment |
add a comment |
$begingroup$
So how does MyFunction know the value of abc
at the delayed function call (when it is called)
if the Private` context isn't on the $ContextPath?
because "CustomPackage`Private`"
is the value of $Context
when MyFunction
is defined (i.e. it is not just $ContextPath
that determines what a function sees but also what is on $Context
).
TL:DR
This is a timely question because it indirectly touches upon the competing imperatives of developers and end-users. To the question itself:
The whole point of packages is that they are a form of encapsulation that allows developers to, without interferance, implement functionality for end-users without bothering them with the underlying details. In particular, the encapsulation involves controlling namespaces so that the underlying details can involve symbols that help implement the functionality but ultimately don't end-up polluting a user's namespace. All symbols defined in a "*`Private`"
namespace are created for exactly this purpose.
Hence in the OP's example, the variable abc
is an underlying detail for the implementation of the public MyFunction
. The developer needs the "detail" of abc
but this particular symbol is of no direct interest to an end-user who typically just ends up calling MyFunction[]
.
The package layout achieves this encapsulation by manipulating $ContextPath
and $Context
as the control-flow passes through the package when it is first loaded. This is described in the other answers and documentation but it can be useful to see it directly:
loc[n_] := Sow[<|
"Location" -> n,
"$Context" -> $Context,
"$ContextPath" -> $ContextPath|>];
Reap[
loc@1;
BeginPackage["CustomPackage`"];
loc@2;
MyFunction::usage = "MyFunction[arg1] adds 5 to arg1.";
Begin["`Private`"];
loc@3;
abc = 5;
MyFunction[arg1_] := arg1 + abc;
End[];
loc@4;
EndPackage[];
loc@5
]// Last // Dataset
When I load the package <the $ContextPath will have CustomPackage on it, but not CustomPackage
Private
Yes, this implements both the public exporting of all CustomPackage
functions but without polluting end-users namespaces with implementation details. In code around Location 3, all packages are cleared out thereby eliminating possible conflicts with existing abc
definitions in currently loaded packages. This is encapsulation benefitting developers but the encapsulation benefitting end-users, as observed, is that on exiting (at Location 5) $ContextPath
contains "CustomPackage`"
(to provide access to MyFunction
) but not "CustomPackage`Private`"
thereby shielding users from symbols used in MyFunction
's implementation.
A programmatic confirmation at Location 5 gives:
MemberQ["CustomPackage`"]@$ContextPath,
MemberQ["CustomPackage`Private`"]@$ContextPath,
Context["abc"]
True, False, "Global`"
At Location 3 in the control-flow, the symbol abc
is not contained in any of the contexts defined in $ContextPath
, ("CustomPackage`"
, or "System`"
) nor is it (yet) in the context defined in $Context
("CustomPackage`Private`"
). Consequently, the name abc
gets created in the context currently set to $Context
. At this location $Context
has value "CustomPackage`Private`"
and hence the symbol CustomPackage`Private`abc
is created. When the control flow then moves on to MyFunction[]
, "CustomPackage`Private`"
is still the value of $Context
so this function "sees" abc
(hence it not just $ContextPath
that determines what a function sees but what is on both$ContextPath
and $Context
).
Note how the convention of placing usage definitions at Location 2 is ostensibly for documentation purposes but its more important role is to ensure that the function goes into the package's context (see $Context
at Location 2) before subsequently being made available in the implementation and for end-users (see $ContextPath
at Locations 3 and 5).
IMO it is kind of cool how these placement protocols just work intuitively without necessarily keeping front-of-mind all the control-flow manipulations, variable-creation mechanisms etc taking place behind the scenes. Hence this means being very careful changing the framework but also IMHO the time is ripe for such extensions given that the line between users/developers may well be in the process of blurring.
$endgroup$
add a comment |
$begingroup$
So how does MyFunction know the value of abc
at the delayed function call (when it is called)
if the Private` context isn't on the $ContextPath?
because "CustomPackage`Private`"
is the value of $Context
when MyFunction
is defined (i.e. it is not just $ContextPath
that determines what a function sees but also what is on $Context
).
TL:DR
This is a timely question because it indirectly touches upon the competing imperatives of developers and end-users. To the question itself:
The whole point of packages is that they are a form of encapsulation that allows developers to, without interferance, implement functionality for end-users without bothering them with the underlying details. In particular, the encapsulation involves controlling namespaces so that the underlying details can involve symbols that help implement the functionality but ultimately don't end-up polluting a user's namespace. All symbols defined in a "*`Private`"
namespace are created for exactly this purpose.
Hence in the OP's example, the variable abc
is an underlying detail for the implementation of the public MyFunction
. The developer needs the "detail" of abc
but this particular symbol is of no direct interest to an end-user who typically just ends up calling MyFunction[]
.
The package layout achieves this encapsulation by manipulating $ContextPath
and $Context
as the control-flow passes through the package when it is first loaded. This is described in the other answers and documentation but it can be useful to see it directly:
loc[n_] := Sow[<|
"Location" -> n,
"$Context" -> $Context,
"$ContextPath" -> $ContextPath|>];
Reap[
loc@1;
BeginPackage["CustomPackage`"];
loc@2;
MyFunction::usage = "MyFunction[arg1] adds 5 to arg1.";
Begin["`Private`"];
loc@3;
abc = 5;
MyFunction[arg1_] := arg1 + abc;
End[];
loc@4;
EndPackage[];
loc@5
]// Last // Dataset
When I load the package <the $ContextPath will have CustomPackage on it, but not CustomPackage
Private
Yes, this implements both the public exporting of all CustomPackage
functions but without polluting end-users namespaces with implementation details. In code around Location 3, all packages are cleared out thereby eliminating possible conflicts with existing abc
definitions in currently loaded packages. This is encapsulation benefitting developers but the encapsulation benefitting end-users, as observed, is that on exiting (at Location 5) $ContextPath
contains "CustomPackage`"
(to provide access to MyFunction
) but not "CustomPackage`Private`"
thereby shielding users from symbols used in MyFunction
's implementation.
A programmatic confirmation at Location 5 gives:
MemberQ["CustomPackage`"]@$ContextPath,
MemberQ["CustomPackage`Private`"]@$ContextPath,
Context["abc"]
True, False, "Global`"
At Location 3 in the control-flow, the symbol abc
is not contained in any of the contexts defined in $ContextPath
, ("CustomPackage`"
, or "System`"
) nor is it (yet) in the context defined in $Context
("CustomPackage`Private`"
). Consequently, the name abc
gets created in the context currently set to $Context
. At this location $Context
has value "CustomPackage`Private`"
and hence the symbol CustomPackage`Private`abc
is created. When the control flow then moves on to MyFunction[]
, "CustomPackage`Private`"
is still the value of $Context
so this function "sees" abc
(hence it not just $ContextPath
that determines what a function sees but what is on both$ContextPath
and $Context
).
Note how the convention of placing usage definitions at Location 2 is ostensibly for documentation purposes but its more important role is to ensure that the function goes into the package's context (see $Context
at Location 2) before subsequently being made available in the implementation and for end-users (see $ContextPath
at Locations 3 and 5).
IMO it is kind of cool how these placement protocols just work intuitively without necessarily keeping front-of-mind all the control-flow manipulations, variable-creation mechanisms etc taking place behind the scenes. Hence this means being very careful changing the framework but also IMHO the time is ripe for such extensions given that the line between users/developers may well be in the process of blurring.
$endgroup$
add a comment |
$begingroup$
So how does MyFunction know the value of abc
at the delayed function call (when it is called)
if the Private` context isn't on the $ContextPath?
because "CustomPackage`Private`"
is the value of $Context
when MyFunction
is defined (i.e. it is not just $ContextPath
that determines what a function sees but also what is on $Context
).
TL:DR
This is a timely question because it indirectly touches upon the competing imperatives of developers and end-users. To the question itself:
The whole point of packages is that they are a form of encapsulation that allows developers to, without interferance, implement functionality for end-users without bothering them with the underlying details. In particular, the encapsulation involves controlling namespaces so that the underlying details can involve symbols that help implement the functionality but ultimately don't end-up polluting a user's namespace. All symbols defined in a "*`Private`"
namespace are created for exactly this purpose.
Hence in the OP's example, the variable abc
is an underlying detail for the implementation of the public MyFunction
. The developer needs the "detail" of abc
but this particular symbol is of no direct interest to an end-user who typically just ends up calling MyFunction[]
.
The package layout achieves this encapsulation by manipulating $ContextPath
and $Context
as the control-flow passes through the package when it is first loaded. This is described in the other answers and documentation but it can be useful to see it directly:
loc[n_] := Sow[<|
"Location" -> n,
"$Context" -> $Context,
"$ContextPath" -> $ContextPath|>];
Reap[
loc@1;
BeginPackage["CustomPackage`"];
loc@2;
MyFunction::usage = "MyFunction[arg1] adds 5 to arg1.";
Begin["`Private`"];
loc@3;
abc = 5;
MyFunction[arg1_] := arg1 + abc;
End[];
loc@4;
EndPackage[];
loc@5
]// Last // Dataset
When I load the package <the $ContextPath will have CustomPackage on it, but not CustomPackage
Private
Yes, this implements both the public exporting of all CustomPackage
functions but without polluting end-users namespaces with implementation details. In code around Location 3, all packages are cleared out thereby eliminating possible conflicts with existing abc
definitions in currently loaded packages. This is encapsulation benefitting developers but the encapsulation benefitting end-users, as observed, is that on exiting (at Location 5) $ContextPath
contains "CustomPackage`"
(to provide access to MyFunction
) but not "CustomPackage`Private`"
thereby shielding users from symbols used in MyFunction
's implementation.
A programmatic confirmation at Location 5 gives:
MemberQ["CustomPackage`"]@$ContextPath,
MemberQ["CustomPackage`Private`"]@$ContextPath,
Context["abc"]
True, False, "Global`"
At Location 3 in the control-flow, the symbol abc
is not contained in any of the contexts defined in $ContextPath
, ("CustomPackage`"
, or "System`"
) nor is it (yet) in the context defined in $Context
("CustomPackage`Private`"
). Consequently, the name abc
gets created in the context currently set to $Context
. At this location $Context
has value "CustomPackage`Private`"
and hence the symbol CustomPackage`Private`abc
is created. When the control flow then moves on to MyFunction[]
, "CustomPackage`Private`"
is still the value of $Context
so this function "sees" abc
(hence it not just $ContextPath
that determines what a function sees but what is on both$ContextPath
and $Context
).
Note how the convention of placing usage definitions at Location 2 is ostensibly for documentation purposes but its more important role is to ensure that the function goes into the package's context (see $Context
at Location 2) before subsequently being made available in the implementation and for end-users (see $ContextPath
at Locations 3 and 5).
IMO it is kind of cool how these placement protocols just work intuitively without necessarily keeping front-of-mind all the control-flow manipulations, variable-creation mechanisms etc taking place behind the scenes. Hence this means being very careful changing the framework but also IMHO the time is ripe for such extensions given that the line between users/developers may well be in the process of blurring.
$endgroup$
So how does MyFunction know the value of abc
at the delayed function call (when it is called)
if the Private` context isn't on the $ContextPath?
because "CustomPackage`Private`"
is the value of $Context
when MyFunction
is defined (i.e. it is not just $ContextPath
that determines what a function sees but also what is on $Context
).
TL:DR
This is a timely question because it indirectly touches upon the competing imperatives of developers and end-users. To the question itself:
The whole point of packages is that they are a form of encapsulation that allows developers to, without interferance, implement functionality for end-users without bothering them with the underlying details. In particular, the encapsulation involves controlling namespaces so that the underlying details can involve symbols that help implement the functionality but ultimately don't end-up polluting a user's namespace. All symbols defined in a "*`Private`"
namespace are created for exactly this purpose.
Hence in the OP's example, the variable abc
is an underlying detail for the implementation of the public MyFunction
. The developer needs the "detail" of abc
but this particular symbol is of no direct interest to an end-user who typically just ends up calling MyFunction[]
.
The package layout achieves this encapsulation by manipulating $ContextPath
and $Context
as the control-flow passes through the package when it is first loaded. This is described in the other answers and documentation but it can be useful to see it directly:
loc[n_] := Sow[<|
"Location" -> n,
"$Context" -> $Context,
"$ContextPath" -> $ContextPath|>];
Reap[
loc@1;
BeginPackage["CustomPackage`"];
loc@2;
MyFunction::usage = "MyFunction[arg1] adds 5 to arg1.";
Begin["`Private`"];
loc@3;
abc = 5;
MyFunction[arg1_] := arg1 + abc;
End[];
loc@4;
EndPackage[];
loc@5
]// Last // Dataset
When I load the package <the $ContextPath will have CustomPackage on it, but not CustomPackage
Private
Yes, this implements both the public exporting of all CustomPackage
functions but without polluting end-users namespaces with implementation details. In code around Location 3, all packages are cleared out thereby eliminating possible conflicts with existing abc
definitions in currently loaded packages. This is encapsulation benefitting developers but the encapsulation benefitting end-users, as observed, is that on exiting (at Location 5) $ContextPath
contains "CustomPackage`"
(to provide access to MyFunction
) but not "CustomPackage`Private`"
thereby shielding users from symbols used in MyFunction
's implementation.
A programmatic confirmation at Location 5 gives:
MemberQ["CustomPackage`"]@$ContextPath,
MemberQ["CustomPackage`Private`"]@$ContextPath,
Context["abc"]
True, False, "Global`"
At Location 3 in the control-flow, the symbol abc
is not contained in any of the contexts defined in $ContextPath
, ("CustomPackage`"
, or "System`"
) nor is it (yet) in the context defined in $Context
("CustomPackage`Private`"
). Consequently, the name abc
gets created in the context currently set to $Context
. At this location $Context
has value "CustomPackage`Private`"
and hence the symbol CustomPackage`Private`abc
is created. When the control flow then moves on to MyFunction[]
, "CustomPackage`Private`"
is still the value of $Context
so this function "sees" abc
(hence it not just $ContextPath
that determines what a function sees but what is on both$ContextPath
and $Context
).
Note how the convention of placing usage definitions at Location 2 is ostensibly for documentation purposes but its more important role is to ensure that the function goes into the package's context (see $Context
at Location 2) before subsequently being made available in the implementation and for end-users (see $ContextPath
at Locations 3 and 5).
IMO it is kind of cool how these placement protocols just work intuitively without necessarily keeping front-of-mind all the control-flow manipulations, variable-creation mechanisms etc taking place behind the scenes. Hence this means being very careful changing the framework but also IMHO the time is ripe for such extensions given that the line between users/developers may well be in the process of blurring.
edited yesterday
answered Apr 11 at 4:18
Ronald MonsonRonald Monson
3,1831634
3,1831634
add a comment |
add a comment |
Thanks for contributing an answer to Mathematica Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
Use MathJax to format equations. MathJax reference.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmathematica.stackexchange.com%2fquestions%2f194963%2fwhat-is-the-accessibility-of-a-packages-private-context-variables%23new-answer', 'question_page');
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown