Parsing for Junos static ipv6 routes

[SLarkworthy]

Adds support for parsing ipv6 static routes. Changes the behavior of routing-options rib to only accept certain patterns for rib name.

Junos docs referenced:
https://www.juniper.net/documentation/us/en/software/junos/static-routing/topics/topic-map/config_junos_routing_table.html#id-example-exporting-specific-routes-from-one-routing-table-into-another-routing-table
https://www.juniper.net/documentation/us/en/software/junos/vpn-l3/topics/topic-map/l3-vpns-routing-instances.html#id-configuring-routing-instances-on-pe-routers-in-vpns__d57160e331

[batfish-bot]

This change is 

[codecov]

Codecov Report

Attention: Patch coverage is 76.92308% with 6 lines in your changes missing coverage. Please review.

Project coverage is 72.72%. Comparing base (ab731e3) to head (2cb421b).
Report is 51 commits behind head on master.

Files with missing lines	Patch %	Lines
...fish/grammar/flatjuniper/ConfigurationBuilder.java	76.92%	6 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #9241      +/-   ##
==========================================
- Coverage   72.73%   72.72%   -0.01%     
==========================================
  Files        3314     3314              
  Lines      170027   170046      +19     
  Branches    20046    20045       -1     
==========================================
+ Hits       123665   123674       +9     
- Misses      37203    37211       +8     
- Partials     9159     9161       +2     

Files with missing lines	Coverage Δ
...fish/grammar/flatjuniper/ConfigurationBuilder.java	69.47% <76.92%> (-0.01%)	⬇️

... and 3 files with indirect coverage changes

[dhalperi]

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_routing_instances.g4 line 340 at r1 (raw file):

;

ro_static6

I might suggest this should be ror6_static:

routing-options rib <somethingv6 like inet6.0> is why I put the 6 there after the r for rib. Then it's just static.

This would imply more for child prefixes.

[dhalperi]

Reviewed 7 of 7 files at r1, all commit messages.
Reviewable status: all files reviewed, 8 unresolved discussions (waiting on @SLarkworthy)

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_common.g4 line 245 at r1 (raw file):

ipv6_prefix: IPV6_PREFIX;

ipv6_address: IPV6_ADDRESS;

isn't this the same as 2 lines up?

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_routing_instances.g4 line 661 at r1 (raw file):

:
    ROUTE
   (

you can get rid of the parens now to clean it up.

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_routing_instances.g4 line 662 at r1 (raw file):

    ROUTE
   (
      prefix = ip_prefix_default_32

note I refer to this one below.

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_routing_instances.g4 line 673 at r1 (raw file):

:
    ROUTE
   (

ditto on parens.

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_routing_instances.g4 line 674 at r1 (raw file):

    ROUTE
   (
       prefix = ipv6_prefix

you may want ipv6_prefix_default_128 like we have ip_prefix_default_32. It's basically prefix or address both allowed, and if address we make it a /128.

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniperLexer.g4 line 2388 at r1 (raw file):

RIB: 'rib'
    {
        switch(lastTokenType()) {

yeah let's chat about this one. Probably totally reasonable, not sure.

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniperLexer.g4 line 3629 at r1 (raw file):

F_RoutingInstanceNameChar
:
   [A-Za-z] | '-'

maybe add a link?

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniperLexer.g4 line 4983 at r1 (raw file):

M_RibName_PERIOD: '.' -> type(PERIOD);
M_RibName_UINT8: F_Uint8 -> type(UINT8), popMode;
M_RibName_WS: F_WhitespaceChar+ -> skip;

hmm this is what I expected would handle those spaces, so what the heck :)

---

projects/batfish/src/test/java/org/batfish/grammar/flatjuniper/FlatJuniperGrammarTest.java line 6504 at r1 (raw file):

            .get("TEST-VRF.inet6.0");
    {
      assertTrue(routesDefault.getStaticRoutes().isEmpty());

since the variable names are different we don't need the scoping {}. That's needed only if you want to prevent leaks out of context.

I might say:

assertThat(c.getMasterLogicalSystem().getRoutingInstances().get("default").getRibs().get(...).getStaticRoutes(),
    isEmpty())
``` or something like that.

The use of `isEmpty()` is nice because it will not just present failure, it will say "expected empty, but got [...]"

[SLarkworthy]

Reviewable status: 1 of 8 files reviewed, 5 unresolved discussions (waiting on @dhalperi)

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_common.g4 line 245 at r1 (raw file):

Previously, dhalperi (Dan Halperin) wrote…

isn't this the same as 2 lines up?

whoops, removed.

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_routing_instances.g4 line 340 at r1 (raw file):

Previously, dhalperi (Dan Halperin) wrote…

I might suggest this should be ror6_static:

routing-options rib <somethingv6 like inet6.0> is why I put the 6 there after the r for rib. Then it's just static.

This would imply more for child prefixes.

done

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_routing_instances.g4 line 661 at r1 (raw file):

Previously, dhalperi (Dan Halperin) wrote…

you can get rid of the parens now to clean it up.

done

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniper_routing_instances.g4 line 674 at r1 (raw file):

Previously, dhalperi (Dan Halperin) wrote…

you may want ipv6_prefix_default_128 like we have ip_prefix_default_32. It's basically prefix or address both allowed, and if address we make it a /128.

I was wondering about that - done.

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniperLexer.g4 line 2388 at r1 (raw file):

Previously, dhalperi (Dan Halperin) wrote…

yeah let's chat about this one. Probably totally reasonable, not sure.

As discussed, changed the lexer so the name is a single token.

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniperLexer.g4 line 3629 at r1 (raw file):

Previously, dhalperi (Dan Halperin) wrote…

maybe add a link?

done

---

projects/batfish/src/test/java/org/batfish/grammar/flatjuniper/FlatJuniperGrammarTest.java line 6504 at r1 (raw file):

Previously, dhalperi (Dan Halperin) wrote…

since the variable names are different we don't need the scoping {}. That's needed only if you want to prevent leaks out of context.

I might say:

assertThat(c.getMasterLogicalSystem().getRoutingInstances().get("default").getRibs().get(...).getStaticRoutes(),
    isEmpty())
``` or something like that.

The use of `isEmpty()` is nice because it will not just present failure, it will say "expected empty, but got [...]"

I couldn't find a matcher for an empty map, but there is one for an empty collection, so I went with making sure the keySet was empty.

Also I know it's weird to factor out the routesDefault/routesTestVrf vars but I'll be adding more tests for each of them once extraction is implemented.

[dhalperi]

Reviewed 7 of 7 files at r2, all commit messages.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @SLarkworthy)

---

projects/batfish/src/main/antlr4/org/batfish/grammar/flatjuniper/FlatJuniperLexer.g4 line 4972 at r2 (raw file):

M_RibName_INET: (F_RoutingInstanceNameChar+ PERIOD)? INET PERIOD UINT8 -> type(INET_RIB_NAME), popMode;
M_RibName_INET6: (F_RoutingInstanceNameChar+ PERIOD)? INET6 PERIOD UINT8 -> type(INET6_RIB_NAME), popMode;
M_RibName_MPLS: MPLS PERIOD UINT8 -> type(MPLS_RIB_NAME), popMode;

why omit VRF names here? I assume they're still relevant.

[dhalperi]

Reviewed 4 of 4 files at r3, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @SLarkworthy)