mirror/stackexchange-dnscontrol
1
0
Fork 0
mirror of https://github.com/StackExchange/dnscontrol.git synced 2024-05-11 05:55:12 +00:00
Code Issues Releases Activity

DOCS: Migrated documentation website to Gitbook (#1809)

Browse Source
This commit is contained in:
Jeffrey Cafferata
2023-01-20 13:56:20 +01:00
committed by GitHub
parent f4388dd6ad
commit f912b15adc
188 changed files with 145405 additions and 4728 deletions
Download Patch File Download Diff File Expand all files Collapse all files

99
documentation/functions/global/D.md Normal file
View File

@@ -0,0 +1,99 @@
---
name: D
parameters:
- name
- registrar
- modifiers...
parameter_types:
name: string
registrar: string
"modifiers...": DomainModifier[]
---
`D` adds a new Domain for DNSControl to manage. The first two arguments are required: the domain name (fully qualified `example.com` without a trailing dot), and the
name of the registrar (as previously declared with [NewRegistrar](NewRegistrar.md)). Any number of additional arguments may be included to add DNS Providers with [DNSProvider](NewDnsProvider.md),
add records with [A](../domain/A.md), [CNAME](../domain/CNAME.md), and so forth, or add metadata.
Modifier arguments are processed according to type as follows:
- A function argument will be called with the domain object as it's only argument. Most of the [built-in modifier functions](https://docs.dnscontrol.org/language-reference/domain-modifiers) return such functions.
- An object argument will be merged into the domain's metadata collection.
- An array argument will have all of it's members evaluated recursively. This allows you to combine multiple common records or modifiers into a variable that can
be used like a macro in multiple domains.
```javascript
var REGISTRAR = NewRegistrar("name.com");
var r53 = NewDnsProvider("R53");
// simple domain
D("example.com", REGISTRAR, DnsProvider(r53),
A("@","1.2.3.4"),
CNAME("test", "foo.example2.com.")
);
// "macro" for records that can be mixed into any zone
var GOOGLE_APPS_DOMAIN_MX = [
MX('@', 1, 'aspmx.l.google.com.'),
MX('@', 5, 'alt1.aspmx.l.google.com.'),
MX('@', 5, 'alt2.aspmx.l.google.com.'),
MX('@', 10, 'alt3.aspmx.l.google.com.'),
MX('@', 10, 'alt4.aspmx.l.google.com.'),
]
D("example.com", REGISTRAR, DnsProvider(r53),
A("@","1.2.3.4"),
CNAME("test", "foo.example2.com."),
GOOGLE_APPS_DOMAIN_MX
);
```
# Split Horizon DNS
DNSControl supports Split Horizon DNS. Simply
define the domain two or more times, each with
their own unique parameters.
To differentiate the different domains, specify the domains as
`domain.tld!tag`, such as `example.com!inside` and
`example.com!outside`.
```javascript
var REG = NewRegistrar("Third-Party");
var DNS_INSIDE = NewDnsProvider("Cloudflare");
var DNS_OUTSIDE = NewDnsProvider("bind");
D("example.com!inside", REG, DnsProvider(DNS_INSIDE),
A("www", "10.10.10.10")
);
D("example.com!outside", REG, DnsProvider(DNS_OUTSIDE),
A("www", "20.20.20.20")
);
D_EXTEND("example.com!inside",
A("internal", "10.99.99.99")
);
```
A domain name without a `!` is assigned a tag that is the empty
string. For example, `example.com` and `example.com!` are equivalent.
However, we strongly recommend against using the empty tag, as it
risks creating confusion. In other words, if you have `domain.tld`
and `domain.tld!external` you now require humans to remember that
`domain.tld` is the external one. I mean... the internal one. You
may have noticed this mistake, but will your coworkers? Will you in
six months? You get the idea.
DNSControl command line flag `--domains` is an exact match. If you
define domains `example.com!george` and `example.com!john` then:
* `--domains=example.com` will not match either domain.
* `--domains='example.com!george'` will match only match the first.
* `--domains='example.com!george',example.com!john` will match both.
{% hint style="info" %}
**NOTE**: The quotes are required if your shell treats `!` as a special
character, which is probably does. If you see an error that mentions
`event not found` you probably forgot the quotes.
{% endhint %}

23
documentation/functions/global/DEFAULTS.md Normal file
View File

@@ -0,0 +1,23 @@
---
name: DEFAULTS
parameters:
- modifiers...
parameter_types:
"modifiers...": RecordModifier[]
---
`DEFAULTS` allows you to declare a set of default arguments to apply to all subsequent domains. Subsequent calls to [D](D.md) will have these
arguments passed as if they were the first modifiers in the argument list.
```javascript
var COMMON = NewDnsProvider("foo");
// we want to create backup zone files for all domains, but not actually register them.
// also create a default TTL
DEFAULTS( DnsProvider(COMMON,0), DefaultTTL(1000));
D("example.com", REGISTRAR, DnsProvider("R53"), A("@","1.2.3.4")); // this domain will have the defaults set.
// clear defaults
DEFAULTS();
D("example2.com", REGISTRAR, DnsProvider("R53"), A("@","1.2.3.4")); // this domain will not have the previous defaults.
```

39
documentation/functions/global/DOMAIN_ELSEWHERE.md Normal file
View File

@@ -0,0 +1,39 @@
---
name: DOMAIN_ELSEWHERE
parameters:
- registrar
- nameserver_names
parameter_types:
registrar: string
nameserver_names: string[]
---
`DOMAIN_ELSEWHERE()` is a helper macro that lets you easily indicate that
a domain's zones are managed elsewhere. That is, it permits you easily delegate
a domain to a hard-coded list of DNS servers.
`DOMAIN_ELSEWHERE` is useful when you control a domain's registrar but not the
DNS servers. For example, suppose you own a domain but the DNS servers are run
by someone else, perhaps a SaaS product you've subscribed to or a DNS server
that is run by your brother-in-law who doesn't trust you with the API keys that
would let you maintain the domain using DNSControl. You need an easy way to
point (delegate) the domain at a specific list of DNS servers.
For example these two statements are equivalent:
```javascript
DOMAIN_ELSEWHERE("example.com", REG_NAMEDOTCOM, ["ns1.foo.com", "ns2.foo.com"]);
// ...is equivalent to...
D("example.com", REG_NAMEDOTCOM,
NO_PURGE,
NAMESERVER("ns1.foo.com"),
NAMESERVER("ns2.foo.com")
);
```
{% hint style="info" %}
**NOTE**: The `NO_PURGE` is used out of abundance of caution but since no
`DnsProvider()` statements exist, no updates would be performed.
{% endhint %}

42
documentation/functions/global/DOMAIN_ELSEWHERE_AUTO.md Normal file
View File

@@ -0,0 +1,42 @@
---
name: DOMAIN_ELSEWHERE_AUTO
parameters:
- domain
- registrar
- dns provider
parameter_types:
domain: string
registrar: string
dns provider: string
---
`DOMAIN_ELSEWHERE_AUTO()` is similar to `DOMAIN_ELSEWHERE()` but instead of
a hardcoded list of nameservers, a DnsProvider() is queried.
`DOMAIN_ELSEWHERE_AUTO` is useful when you control a domain's registrar but the
DNS zones are managed by another system. Luckily you have enough access to that
other system that you can query it to determine the zone's nameservers.
For example, suppose you own a domain but the DNS servers for it are in Azure.
Further suppose that something in Azure maintains the zones (automatic or
human). Azure picks the nameservers for the domains automatically, and that
list may change occasionally. `DOMAIN_ELSEWHERE_AUTO` allows you to easily
query Azure to determine the domain's delegations so that you do not need to
hard-code them in your dnsconfig.js file.
For example these two statements are equivalent:
```javascript
DOMAIN_ELSEWHERE_AUTO("example.com", REG_NAMEDOTCOM, DSP_AZURE);
// ...is equivalent to...
D("example.com", REG_NAMEDOTCOM,
NO_PURGE,
DnsProvider(DSP_AZURE)
);
```
{% hint style="info" %}
**NOTE**: The `NO_PURGE` is used to prevent DNSControl from changing the records.
{% endhint %}

87
documentation/functions/global/D_EXTEND.md Normal file
View File

@@ -0,0 +1,87 @@
---
name: D_EXTEND
parameters:
- name
- modifiers...
parameter_types:
name: string
"modifiers...": DomainModifier[]
---
`D_EXTEND` adds records (and metadata) to a domain previously defined
by `D()`. It can also be used to add subdomain records (and metadata)
to a previously defined domain.
The first argument is a domain name. If it exactly matches a
previously defined domain, `D_EXTEND()` behaves the same as `D()`,
simply adding records as if they had been specified in the original
`D()`.
If the domain name does not match an existing domain, but could be a
(non-delegated) subdomain of an existing domain, the new records (and
metadata) are added with the subdomain part appended to all record
names (labels), and targets (as appropriate). See the examples below.
Matching the domain name to previously-defined domains is done using a
`longest match` algorithm. If `domain.tld` and `sub.domain.tld` are
defined as separate domains via separate `D()` statements, then
`D_EXTEND('sub.sub.domain.tld', ...)` would match `sub.domain.tld`,
not `domain.tld`.
Some operators only act on an apex domain (e.g.
`CF_REDIRECT` and `CF_TEMP_REDIRECT`). Using them
in a `D_EXTEND` subdomain may not be what you expect.
```javascript
D("domain.tld", REG, DnsProvider(DNS),
A("@", "127.0.0.1"), // domain.tld
A("www", "127.0.0.2"), // www.domain.tld
CNAME("a", "b") // a.domain.tld -> b.domain.tld
);
D_EXTEND("domain.tld",
A("aaa", "127.0.0.3"), // aaa.domain.tld
CNAME("c", "d") // c.domain.tld -> d.domain.tld
);
D_EXTEND("sub.domain.tld",
A("bbb", "127.0.0.4"), // bbb.sub.domain.tld
A("ccc", "127.0.0.5"), // ccc.sub.domain.tld
CNAME("e", "f") // e.sub.domain.tld -> f.sub.domain.tld
);
D_EXTEND("sub.sub.domain.tld",
A("ddd", "127.0.0.6"), // ddd.sub.sub.domain.tld
CNAME("g", "h") // g.sub.sub.domain.tld -> h.sub.sub.domain.tld
);
D_EXTEND("sub.domain.tld",
A("@", "127.0.0.7"), // sub.domain.tld
CNAME("i", "j") // i.sub.domain.tld -> j.sub.domain.tld
);
```
This will end up in the following modifications: (This output assumes the `--verbose` flag)
```text
******************** Domain: domain.tld
----- Getting nameservers from: cloudflare
----- DNS Provider: cloudflare...7 corrections
#1: CREATE A aaa.domain.tld 127.0.0.3
#2: CREATE A bbb.sub.domain.tld 127.0.0.4
#3: CREATE A ccc.sub.domain.tld 127.0.0.5
#4: CREATE A ddd.sub.sub.domain.tld 127.0.0.6
#5: CREATE A sub.domain.tld 127.0.0.7
#6: CREATE A www.domain.tld 127.0.0.2
#7: CREATE A domain.tld 127.0.0.1
#8: CREATE CNAME a.domain.tld b.domain.tld.
#9: CREATE CNAME c.domain.tld d.domain.tld.
#10: CREATE CNAME e.sub.domain.tld f.sub.domain.tld.
#11: CREATE CNAME g.sub.sub.domain.tld h.sub.sub.domain.tld.
#12: CREATE CNAME i.sub.domain.tld j.sub.domain.tld.
```
ProTips: `D_EXTEND()` permits you to create very complex and
sophisticated configurations, but you shouldn't. Be nice to the next
person that edits the file, who may not be as expert as yourself.
Enhance readability by putting any `D_EXTEND()` statements immediately
after the original `D()`, like in above example. Avoid the temptation
to obscure the addition of records to existing domains with randomly
placed `D_EXTEND()` statements. Don't build up a domain using loops of
`D_EXTEND()` statements. You'll be glad you didn't.

44
documentation/functions/global/FETCH.md Normal file
View File

@@ -0,0 +1,44 @@
---
name: FETCH
parameters:
- url
- args
ts_ignore: true
# Make sure to update fetch.d.ts if changing the docs below!
---
`FETCH` is a wrapper for the [Fetch API](https://developer.mozilla.org/en-US/docs/Web/API/Fetch_API). This allows dynamically setting DNS records based on an external data source, e.g. the API of your cloud provider.
Compared to `fetch` from Fetch API, `FETCH` will call [PANIC](PANIC.md) to terminate the execution of the script, and therefore DNSControl, if a network error occurs.
Otherwise the syntax of `FETCH` is the same as `fetch`.
`FETCH` is not enabled by default. Please read the warnings below.
> WARNING:
>
> 1. Relying on external sources adds a point of failure. If the external source doesn't work, your script won't either. Please make sure you are aware of the consequences.
> 2. Make sure DNSControl only uses verified configuration if you want to use `FETCH`. For example, an attacker can send Pull Requests to your config repo, and have your CI test malicious configurations and make arbitrary HTTP requests. Therefore, `FETCH` must be explicitly enabled with flag `--allow-fetch` on DNSControl invocation.
```javascript
var REG_NONE = NewRegistrar('none');
var DNS_BIND = NewDnsProvider('bind');
D('example.com', REG_NONE, DnsProvider(DNS_BIND), [
A('@', '1.2.3.4'),
]);
FETCH('https://example.com', {
// All three options below are optional
headers: {"X-Authentication": "barfoo"},
method: "POST",
body: "Hello World",
}).then(function(r) {
return r.text();
}).then(function(t) {
// Example of generating record based on response
D_EXTEND('example.com', [
TXT('@', t.slice(0, 100)),
]);
});
```

29
documentation/functions/global/IP.md Normal file
View File

@@ -0,0 +1,29 @@
---
name: IP
parameters:
- ip
parameter_types:
ip: string
return: number
---
Converts an IPv4 address from string to an integer. This allows performing mathematical operations with the IP address.
```javascript
var addrA = IP('1.2.3.4')
var addrB = addrA + 1
// addrB = 1.2.3.5
```
{% hint style="info" %}
**NOTE**: `IP()` does not accept IPv6 addresses (PRs gladly accepted!). IPv6 addresses are simply strings:
{% endhint %}
```javascript
// IPv4 Var
var addrA1 = IP("1.2.3.4");
var addrA2 = "1.2.3.4";
// IPv6 Var
var addrAAAA = "0:0:0:0:0:0:0:0";
```

46
documentation/functions/global/NewDnsProvider.md Normal file
View File

@@ -0,0 +1,46 @@
---
name: NewDnsProvider
parameters:
- name
- type
- meta
parameter_types:
name: string
type: string?
meta: object?
return: string
---
NewDnsProvider activates a DNS Service Provider (DSP) specified in `creds.json`.
A DSP stores a DNS zone's records and provides DNS service for the zone (i.e.
answers on port 53 to queries related to the zone).
* `name` must match the name of an entry in `creds.json`.
* `type` specifies a valid DNS provider type identifier listed on the [provider page](../../providers.md).
* Starting with [v3.16](../../v316.md), the type is optional. If it is absent, the `TYPE` field in `creds.json` is used instead. You can leave it out. (Thanks to JavaScript magic, you can leave it out even when there are more fields).
* Starting with v4.0, specifying the type may be an error. Please add the `TYPE` field to `creds.json` and remove this parameter from `dnsconfig.js` to prepare.
* `meta` is a way to send additional parameters to the provider. It is optional and only certain providers use it. See the [individual provider docs](../../providers.md) for details.
This function will return an opaque string that should be assigned to a variable name for use in [D](D.md) directives.
Prior to [v3.16](../../v316.md):
```javascript
var REG_MYNDC = NewRegistrar("mynamedotcom", "NAMEDOTCOM");
var DNS_MYAWS = NewDnsProvider("myaws", "ROUTE53");
D("example.com", REG_MYNDC, DnsProvider(DNS_MYAWS),
A("@","1.2.3.4")
);
```
In [v3.16](../../v316.md) and later:
```javascript
var REG_MYNDC = NewRegistrar("mynamedotcom");
var DNS_MYAWS = NewDnsProvider("myaws");
D("example.com", REG_MYNDC, DnsProvider(DNS_MYAWS),
A("@","1.2.3.4")
);
```

46
documentation/functions/global/NewRegistrar.md Normal file
View File

@@ -0,0 +1,46 @@
---
name: NewRegistrar
parameters:
- name
- type
- meta
parameter_types:
name: string
type: string?
meta: object?
return: string
---
NewRegistrar activates a Registrar Provider specified in `creds.json`.
A registrar maintains the domain's registration and delegation (i.e. the
nameservers for the domain). DNSControl only manages the delegation.
* `name` must match the name of an entry in `creds.json`.
* `type` specifies a valid DNS provider type identifier listed on the [provider page](../../providers.md).
* Starting with [v3.16](../../v316.md), the type is optional. If it is absent, the `TYPE` field in `creds.json` is used instead. You can leave it out. (Thanks to JavaScript magic, you can leave it out even when there are more fields).
* Starting with v4.0, specifying the type may be an error. Please add the `TYPE` field to `creds.json` and remove this parameter from `dnsconfig.js` to prepare.
* `meta` is a way to send additional parameters to the provider. It is optional and only certain providers use it. See the [individual provider docs](../../providers.md) for details.
This function will return an opaque string that should be assigned to a variable name for use in [D](D.md) directives.
Prior to [v3.16](../../v316.md):
```javascript
var REG_MYNDC = NewRegistrar("mynamedotcom", "NAMEDOTCOM");
var DNS_MYAWS = NewDnsProvider("myaws", "ROUTE53");
D("example.com", REG_MYNDC, DnsProvider(DNS_MYAWS),
A("@","1.2.3.4")
);
```
In [v3.16](../../v316.md) and later:
```javascript
var REG_MYNDC = NewRegistrar("mynamedotcom");
var DNS_MYAWS = NewDnsProvider("myaws");
D("example.com", REG_MYNDC, DnsProvider(DNS_MYAWS),
A("@","1.2.3.4")
);
```

14
documentation/functions/global/PANIC.md Normal file
View File

@@ -0,0 +1,14 @@
---
name: PANIC
parameters:
- message
parameter_types:
message: string
ts_return: never
---
`PANIC` terminates the script and therefore DNSControl with an exit code of 1. This should be used if your script cannot gather enough information to generate records, for example when a HTTP request failed.
```javascript
PANIC("Something really bad has happened");
```

54
documentation/functions/global/REV.md Normal file
View File

@@ -0,0 +1,54 @@
---
name: REV
parameters:
- address
parameter_types:
address: string
ts_return: string
---
`REV` returns the reverse lookup domain for an IP network. For
example `REV('1.2.3.0/24')` returns `3.2.1.in-addr.arpa.` and
`REV('2001:db8:302::/48)` returns `2.0.3.0.8.b.d.0.1.0.0.2.ip6.arpa.`.
This is used in `D()` functions to create reverse DNS lookup zones.
This is a convenience function. You could specify `D('3.2.1.in-addr.arpa',
...` if you like to do things manually but why would you risk making
typos?
`REV` complies with RFC2317, "Classless in-addr.arpa delegation"
for netmasks of size /25 through /31.
While the RFC permits any format, we abide by the recommended format:
`FIRST/MASK.C.B.A.in-addr.arpa` where `FIRST` is the first IP address
of the zone, `MASK` is the netmask of the zone (25-31 inclusive),
and A, B, C are the first 3 octets of the IP address. For example
`172.20.18.130/27` is located in a zone named
`128/27.18.20.172.in-addr.arpa`
If the address does not include a "/" then `REV` assumes /32 for IPv4 addresses
and /128 for IPv6 addresses.
Note that the lower bits (the ones outside the netmask) must be zeros. They are not
zeroed out automatically. Thus, `REV('1.2.3.4/24')` is an error. This is done
to catch typos.
```javascript
D(REV('1.2.3.0/24'), REGISTRAR, DnsProvider(BIND),
PTR("1", 'foo.example.com.'),
PTR("2", 'bar.example.com.'),
PTR("3", 'baz.example.com.'),
// These take advantage of DNSControl's ability to generate the right name:
PTR("1.2.3.10", 'ten.example.com.'),
);
D(REV('2001:db8:302::/48'), REGISTRAR, DnsProvider(BIND),
PTR("1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0", 'foo.example.com.'), // 2001:db8:302::1
// These take advantage of DNSControl's ability to generate the right name:
PTR("2001:db8:302::2", 'two.example.com.'), // 2.0.0...
PTR("2001:db8:302::3", 'three.example.com.'), // 3.0.0...
);
```
In the future we plan on adding a flag to `A()` which will insert
the correct PTR() record if the appropriate `D(REV()` domain (i.e. `.arpa` domain) has been
defined.

62
documentation/functions/global/getConfiguredDomains.md Normal file
View File

@@ -0,0 +1,62 @@
---
name: getConfiguredDomains
ts_is_function: true
ts_return: string[]
---
`getConfiguredDomains` getConfiguredDomains is a helper function that returns the domain names
configured at the time the function is called. Calling this function early or later in
`dnsconfig.js` may return different results. Typical usage is to iterate over all
domains at the end of your configuration file.
Example for adding records to all configured domains:
```javascript
var domains = getConfiguredDomains();
for(i = 0; i < domains.length; i++) {
D_EXTEND(domains[i],
TXT('_important', 'BLA') // I know, not really creative.
)
}
```
This will end up in following modifications: (All output assumes the `--verbose` flag)
```text
******************** Domain: domain1.tld
----- Getting nameservers from: registrar
----- DNS Provider: registrar...2 corrections
#1: CREATE TXT _important.domain1.tld "BLA" ttl=43200
#2: REFRESH zone domain1.tld
******************** Domain: domain2.tld
----- Getting nameservers from: registrar
----- DNS Provider: registrar...2 corrections
#1: CREATE TXT _important.domain2.tld "BLA" ttl=43200
#2: REFRESH zone domain2.tld
```
Example for adding DMARC report records:
This example might be more useful, specially for configuring the DMARC report records. According to DMARC RFC you need to specify `domain2.tld._report.dmarc.domain1.tld` to allow `domain2.tld` to send aggregate/forensic email reports to `domain1.tld`. This can be used to do this in an easy way, without using the wildcard from the RFC.
```javascript
var domains = getConfiguredDomains();
for(i = 0; i < domains.length; i++) {
D_EXTEND("domain1.tld",
TXT(domains[i] + '._report._dmarc', 'v=DMARC1')
);
}
```
This will end up in following modifications:
```text
******************** Domain: domain2.tld
----- Getting nameservers from: registrar
----- DNS Provider: registrar...4 corrections
#1: CREATE TXT domain1.tld._report._dmarc.domain2.tld "v=DMARC1" ttl=43200
#2: CREATE TXT domain3.tld._report._dmarc.domain2.tld "v=DMARC1" ttl=43200
#3: CREATE TXT domain4.tld._report._dmarc.domain2.tld "v=DMARC1" ttl=43200
#4: REFRESH zone domain2.tld
```

92
documentation/functions/global/require.md Normal file
View File

@@ -0,0 +1,92 @@
---
name: require
parameters:
- path
ts_ignore: true
---
`require(...)` loads the specified JavaScript or JSON file, allowing
to split your configuration across multiple files.
If the supplied `path` string ends with `.js`, the file is interpreted
as JavaScript code, almost as though its contents had been included in
the currently-executing file. If the path string ends with `.json`,
`require()` returns the `JSON.parse()` of the file's contents.
If the path string begins with a `.`, it is interpreted relative to
the currently-loading file (which may not be the file where the
`require()` statement is, if called within a function), otherwise it
is interpreted relative to the program's working directory at the time
of the call.
```javascript
// dnsconfig.js
require('kubernetes/clusters.js');
D("mydomain.net", REG, PROVIDER,
IncludeKubernetes()
);
```
```javascript
// kubernetes/clusters.js
require('./clusters/prod.js');
require('./clusters/dev.js');
function IncludeKubernetes() {
return [includeK8Sprod(), includeK8Sdev()];
}
```
```javascript
// kubernetes/clusters/prod.js
function includeK8Sprod() {
return [
// ...
];
}
```
```javascript
// kubernetes/clusters/dev.js
function includeK8Sdev() {
return [
// ...
];
}
```
You can also use it to require JSON files and initialize variables with it:
```javascript
// dnsconfig.js
var domains = require('./domain-ip-map.json')
for (var domain in domains) {
D(domain, REG, PROVIDER,
A("@", domains[domain])
);
}
```
```javascript
// domain-ip-map.json
{
"mydomain.net": "1.1.1.1",
"myotherdomain.org": "5.5.5.5"
}
```
# Future
It might be better to rename the function to something like
`include()` instead, (leaving `require` as a deprecated alias) because
by analogy it is *much* closer to PHP's `include()` function than it
is to node's `require()`. After all, the reason node.js calls it
"require" is because it's a declarative statement saying the file is
needed, and so should be loaded if it hasn't already been loaded.
In contrast, dnscontrol's require is actually an imperative command to
load the file and execute the code or parse the data from it. (So if
two files both `require("./tools.js")`, for example, then it will be
loaded twice, whereas in node.js it would only be loaded once.)

46
documentation/functions/global/require_glob.md Normal file
View File

@@ -0,0 +1,46 @@
---
name: require_glob
parameters:
- path
- recursive
parameter_types:
path: string
recursive: boolean
---
`require_glob()` can recursively load `.js` files, optionally non-recursive as well.
Possible parameters are:
- Path as string, where you would like to start including files. Mandatory. Pattern matching possible, see [GoLand path/filepath/#Match docs](https://golang.org/pkg/path/filepath/#Match).
- If being recursive. This is a boolean if the search should be recursive or not. Define either `true` or `false`. Default is `true`.
Example to load `.js` files recursively:
```javascript
require_glob("./domains/");
```
Example to load `.js` files only in `domains/`:
```javascript
require_glob("./domains/", false);
```
One more important thing to note: `require_glob()` is as smart as `require()` is. It loads files always relative to the JavaScript
file where it's being executed in. Let's go with an example, as it describes it better:
`dnscontrol.js`:
```javascript
require("domains/index.js");
```
`domains/index.js`:
```javascript
require_glob("./user1/");
```
This will now load files being present underneath `./domains/user1/` and **NOT** at below `./domains/`, as `require_glob()`
is called in the subfolder `domains/`.